Promotion and constraint of adaptive evolution in cave-dwelling lineages
Hilary Allison Edgington Sharon Center, Ohio
Bachelor of Arts, The College of Wooster, 2010
A Dissertation presented to the Graduate Faculty of the University of Virginia in Candidacy for the Degree of Doctor of Philosophy
Department of Biology
University of Virginia May, 2015
i
Dissertation Abstract
Experiencing a novel environment is a common driver of evolutionary change.
This change can be promoted by a variety of mechanisms and outcomes, such as novel selection, decreases in genetic variation due to founder effects, drift effects, and genetic admixture. Equally important are forces that impede or prevent change, including lack of suitable mutations, genetic correlations, developmental constraints, or antagonistic selection. A valuable system for studying questions related to evolutionary change following colonization of a novel habitat is the cave habitat, which has induced dramatic adaptive changes in many species that inhabit caves as a result of both adaptive and neutral forces. I studied the evolution of cave-dwelling salamanders, and in particular in the cave salamander, Eurycea lucifuga. Eurycea lucifuga is an evolutionary puzzle, inhabiting caves while having maintained brilliant coloration more typical of ancestral, surface dwelling ancestors.
I first compared body shape among salamanders within Eurycea to examine the prediction from many previous studies that cave-dwellers are elongated relative to non- cave-dwellers. I found that terrestrial species were larger than aquatic species, and also that cave species had shorter tails relative to their body size than non-cave species. These results contrast with trends generally discussed in the cave literature, and reflect a need for explicit testing of how habitat impacts morphology in salamanders. I next investigated the phylogeographic history of Eurycea lucifuga and found that divergence among major lineages happened millions of years ago, with more recent expansion within each clade. This is similar to other cave-dwellers that show extreme morphological specialization to the subterranean climate. Our results do not support the idea that a lack ii of time underground acts as a constraint on adaptive evolution in this species. Lastly, I compared phenotypic differentiation in color traits with neutral genetic differentiation.
The findings indicated that at small scales there is evidence of local differentiation in color relative to the neutral expectation. However, the spatial structuring of differentiation differs between the color phenotype and genotype. Population genetic analyses within a cave system indicated that migration likely occurs by surface corridors, suggesting that coloration may be maintained by a more substantial reliance on non-cave habitats than has been documented for this species.
I concluded from these results that the phenotype of the cave salamander, Eurycea lucifuga, is greatly impacted by a relatively minor aspect of its ecology, and hypothesize that its non-cave morphology is selectively maintained by use of surface habitats. In general, this work emphasizes the need to examine trait change in a broad context, considering phylogenetics, ecology, and neutral processes. iii
Acknowledgements
Though it is impossible to name all of the people who have contributed to the completion of this dissertation, there are a few who require special recognition. First and foremost, I am deeply grateful to my advisor, Doug Taylor, who has contributed overwhelmingly to any professional and personal growth I experienced during my time in his lab. From my first semester in the department, Doug put a great deal of energy into helping me identify my proximate and ultimate goals, and then supporting my pursuit of them wholeheartedly. His enthusiasm for unraveling scientific puzzles and the care he has put into mentoring me are examples I hope to follow someday in my own lab. I am also indebted to my committee members, Bob Cox, Butch Brodie, Laura Galloway, and
Dave Carr. Their guidance and support, as well as their critical evaluation of the design, execution, and presentation of these projects, have been invaluable, and I could not have asked for a better group of mentors. Bob Cox in particular has been an incredibly conscientious first reader, and consistently went above and beyond to ensure that I was on track, both scientifically and administratively. I greatly appreciate his patience and keen insight.
There exists a group of people who, like the Cave Salamander, display an overwhelming preference for the subterranean habitat. None of the research I have conducted would have been possible if not for the assistance of cavers across the country, who gave up their time to assist me in accessing caves and collecting samples. Though I will never attain their level of enthusiasm for caving, being deep underground with these fine folks has made me acutely aware of how incredible the natural world is, and what a small part of it I inhabit. I am also very grateful for the assistance of wildlife iv professionals throughout the country for their assistance with cave access and permit processing.
I have been fortunate to work in a lab composed of exceptionally talented members, past and present. This project would not exist were it not for Peter Fields who first introduced me to the Cave Salamander at trivia night five years ago, and whose patience with my ignorant questions has been truly admirable over the years. Colleen
Ingram came along just as I was starting to know enough about phylogenetics to get into trouble with it, and aside from her invaluable help in that regard she has proved to be a great friend and collaborator. I owe a great deal to Andrea Berardi, who has been a wonderful friend and a great example for me during her time in the Taylor lab and beyond. I am particularly thankful for her assistance in developing the ideas for Chapter
3, as well as her keen editing eye and insightful comments.
This effort would have been abandoned long ago were it not for the feedback, suggestions, distractions, and coaching from Corlett Wood, Brian Sanderson, Malcolm
Augat, and Karen Barnard-Kubow. Their friendship has enriched my life in
Charlottesville more than I can express, and I am excited to find out what amazing directions their lives will all take in the future.
Finally, the amount of support I have received throughout this process from my parents, my sister, and her family continues to amaze and humble me. The value my parents place on being a life-long learner and their pursuit of that goal in their own lives has been a constant inspiration to me, and I could not have asked for a better environment in which to start my exploration of the natural world. I am incredibly grateful that in moving to Charlottesville I was able to move closer to my sister Margaret as her family v has grown and changed. Not only has Margaret been a constant example for me in her selfless concern for each person in her life and the strength she displays during turbulent times, she also spent a weekend caving with me at great mental and physical risk.
Thank you all.
vi
TABLE OF CONTENTS
Abstract……………………………...………………………………………. i Acknowledgements………………………………..………………………… iii Table of Contents…………………………………………….……………… vi
Introduction……………………………………..…………………………… 1
Chapter 1: The phylogenetic history of body shape variation in salamanders of the Plethodontid genus Eurycea (Plethodontidae)…………….………….. 15
Figures and Tables…………………………………………...………. 33
Chapter 2: The phylogeography and population genetics of a cave-dwelling salamander, Eurycea lucifuga………………………………….…………….. 56
Figures and Tables………………………………...…………………. 84
Chapter 3: Varying color differentiation at multiple spatial scales in the cave salamander, Eurycea lucifuga (Plethodontidae)…...………………...……….. 97
Figures and Tables……………………………………………………. 130
Appendix A: Development and characterization of 19 microsatellite loci in the cave salamander, Eurycea lucifuga (Plethodontidae)….……………………. 145
1 Introduction
Understanding the parameters that impact a species’ ability to adapt and survive in a novel environment has been crucial to evolutionary biology, conservation biology, and ecology (Bradshaw 1991; Charlesworth & Hughes 2000; Reed et al. 2003). Exposure to novel environments can come about primarily two ways: either a change occurs in a population’s current range, or there is dispersal and colonization of a new space. Perhaps the most ubiquitous example of these processes is the disruption of environment that accompanied glaciation in the Pleistocene, which not only altered the physiology and behavior of many species, but also dramatically impacted many species’ ranges (Avise
2000). There are many features of both populations and habitats that will make colonization more or less likely. Colonization success varies across species with differing biotic traits such as taxon, species, body size, and functional group (Eggleston et al. 1999), and is more likely in species that display omnivory, gregariousness, and asexuality (Ehrlich 1989; Lodge 1993). Features such as the type and size of a species’ abiotic habitat also impact colonization success (Eggleston et al. 1999). It has been suggested that some colonizers gain a benefit from preadaptation, for example an ability to use the widest range or commonest type of habitat (Thomas et al. 2001), or from phenotypic plasticity (Baldwin 1896; Ehrlich 1989; Holway & Suarez 1999).
Once a novel environment has been experienced, populations can change in a variety of ways. The most obvious mechanism through which environment causes phenotypic change is that populations are exposed to new selection pressures, with simultaneous relaxation of selection from the previous environment (Mitchell & Power
2003; Torchin et al. 2003). A well-known example of adaptive morphological change is 2 the industrial melanism observed in the Peppered Moth (Cook et al. 1986; Kettlewell
1961), where the introduction of a new selection pressure (darkening of the trees due to pollution) caused a change in pigment to be beneficial for certain populations.
Colonization can also impact the amount of genetic variation available for selection to work with. Dispersal to a new habitat is often accompanied by genetic bottleneck events due to the relatively small number of individuals in the new population (Chakraborty &
Nei 1977; Nei et al. 1975). This affects adaptive evolution not only because of the composition of the founder population, but also because of its vulnerability to stochastic changes (Santos et al. 2012). Neutral forces such as drift and admixture may also cause phenotypic and genetic changes in founder populations (Clegg et al. 2002). Drift was the primary force behind geographic genetic and morphometric variation in the Common
Chaffinch, Fringilla coelebs, which has led to both intra- and interspecific divergence
(Baker et al. 1990). Variation in lamella number and body size in introduced populations of the Brown Anole, Anolis sagrei was found to be a result of admixture between different source populations and not of natural selection (Kolbe et al. 2004; Kolbe et al.
2007). Other factors that may affect genetic variation of colonizing populations include the mutation rate and patterns of dominance, epistasis, and pleiotropy of mutations (Reed et al. 2003).
Often we think about ways that evolutionary change can be promoted in colonizing populations, but equally important to the evolutionary trajectory can be constraining factors on adaptive change. Futuyma (2010) outlines several limits on response to selection: First, there may be a lack of suitable mutations on which selection may act, or the mutation rate may be slow enough to limit genetic variation. Next, a 3 character may be genetically correlated with another character under antagonistic selection. There may also be a lack of genetic correlation among traits, which could make selection in a particular direction ineffective if the benefits of a phenotype require change in multiple traits (Futuyma 2010). Gene flow among populations can also act as a constraint when gene swamping overwhelms rare alleles, particularly in small marginal populations, although the effects of gene flow will vary depending on whether the migrating alleles are beneficial, neutral, or deleterious (Antonovics 1976). Limits to adaptation also come from development, since the structure, character composition, or dynamics of the developmental system may not allow for certain changes (Smith et al.
1985). For example, a constraint local to Palms involves a lack of secondary thickening in trunk tissue, which prevents branching structures from being successful in that taxon.
A more universal constraint involves a skeletal trade-off, where change to favor speed necessarily causes a decrease in potential for applied force (Smith et al. 1985).
Developmental constraints can generally be identified using measurements of selection and heritability, or comparisons among taxa. A further constraint involves the idea that adaptive evolution requiring multiple changes can only happen if the intermediate steps are beneficial, or at least non-deleterious (Smith et al. 1985).
There may be benefits to populations that avoid adapting completely to one environment, although evidence for this has been mixed. The jack-of-all-trades, master of none hypothesis (Ehinger et al. 2014; Huey & Hertz 1984) predicts that specialization to a resource will increase a population’s fitness in that environment, but will cause a trade-off in the population’s fitness in other environments. This hypothesis has been supported by findings that inhabiting spatially or temporally heterogeneous environment 4 results in more genetic and phenotypic diversity (Buckling & Rainey 2002; Porter &
Rice 2013; Reed et al. 2003; Travisano et al. 1995) (but see (Hawthorne 1997)) and that specialists obtained a greater benefit from their symbionts than did generalists (Ehinger et al. 2014). This view of specialists and generalists also supports the niche width-variation hypothesis, which predicts that species with more or broader niches should be more polymorphic (Soulé 1971). There is also evidence that populations inhabiting variable environments are more successful when faced with a novel stressful environment (Reed et al. 2003). The limitations imposed by adaptive specialization seem to vary among taxa. In a comparison of generalist and specialist lady beetles in the genus Harmonia
(Coccinelidae), similar fitness measurements of both species raised on the resource of the specialist indicated that host range in these species is not motivated by intrinsic suitability
(Noriyuki & Osawa 2012). Lack of evidence for performance trade-offs on different hosts was also found in other systems (Agosta & Klemens 2009; Futuyma 2008), which suggests that specialization may evolve through multiple mechanisms. The consequences of specialization seem to vary also in different systems: while specialization within some species in the angiosperm genus Ruellia (Acanthacaea) (for example, pollination by hawkmoths or bats) have lead to evolutionary ‘dead-ends’, limiting the amount of further diversification, others (such as hummingbird pollination) have not (Tripp & Manos
2008).
Study System
Cave communities have been the focus of a large body of literature investigating interactions between environment and phenotype because of the repeated evolution of a consistent collection of traits known as troglomorphy across many taxa. Troglomorphy 5 includes such phenotypic change as reduced eyes and blindness, limb and body elongation, depigmentation, reduced metabolism, slower life histories, and paedomorphosis (Porter 2007). Caves represent a relatively simple habitat in terms of environmental consistency, selection pressure, and community structure (Barr &
Holsinger 1985; Howarth 1993), and the mechanism behind phenotypic change in cave- dwellers is well characterized (Jeffery 2005; Jeffery 2009; Protas et al. 2007; Protas et al.
2011; Yamamoto & Jeffery 2000; Yamamoto et al. 2004). There is debate surrounding the process of colonization, which tends to center around two alternative hypotheses: exploratory use of caves by some populations, which are then isolated when climatic change causes extinction events in surface dwelling populations (Barr & Holsinger 1985), or sympatric isolation between surface and cave populations caused by adaptive divergence in the cave-dwellers (Howarth 1973). Regardless of the causal order of phenotypic and genetic divergence between surface and cave populations, troglomorphy is generally viewed as a relatively simple one-way path to adaptive specialization resulting in the restriction of taxa to a subterranean habitat.
Cave-restricted individuals, or troglobites, often share the cave habitat with other classes of cave-dwellers with different ecological and phenotypic characteristics.
Troglophiles are a class of cave species that establish permanent populations in caves, but are not restricted to that habitat (Sket 2008). These species lack many or most of the traits characterized by troglomorphy, and for many years were thought to represent younger lineages acting as a stepping stone between surface-dweller and cave-relict (Sket
2008). However, genetic studies have shown no clear correlation between age of a cave lineage and its degree of troglomorphy (Wessel et al. 2007). Although troglophiles tend 6 to exhibit more gene flow between populations than troglobites (Porter 2007), the degree to which they are able to traverse non-cave environments varies greatly depending on characteristics of both the landscape and the species (Caccone 1985).
Most of this dissertation focuses on a troglophilic salamander, Eurycea lucifuga, which is a common inhabitant of karstic caves across the central and Eastern United
States. Cave-dwelling has evolved independently at least five times within Eurycea
(Bonett et al. 2014), and the genus includes the greatest number of troglobites of any vertebrate taxon (Culver et al. 2000). While E. lucifuga is restricted to caves during the egg and larval stages of its life cycle, adults are occasionally found outside of caves,
(Hutchison 1958; Petranka 1998). Despite its ubiquity in the cave habitat and the contrast between its habitat and phenotype, little is known about the evolutionary history or ecology of Eurycea lucifuga.
Purpose
My goal in this body of work was to examine how the evolutionary history and ecology of Eurycea lucifuga have led to its phenotype, given its divergent evolutionary trajectory from that of closely related cave-dwellers. More specifically, I aimed to examine potential reasons why many species, after colonizing the cave environment, gain troglomorphic characteristics and become restricted to caves, and yet Eurycea lucifuga and other troglophilic species retained the ancestral phenotype following colonization of caves. I formed two hypotheses, which are not mutually exclusive: that this species represents a very recent colonization of the cave habitat, or that maintenance of the ancestral phenotype is a result of antagonistic selection due to its use of the surface habitat. This dissertation is composed of three chapters and an Appendix. 7 Chapter 1: I performed morphometric comparisons among terrestrial and aquatic, and among cave and non-cave species within a broad sampling of salamander species. These comparisons were performed using standard generalized linear mixed models, as well as phylogenetic ANOVA. I was specifically interested in testing a long-held hypothesis that elongation increases following cave colonization. Results indicate that the majority of variation among habitat groups is found between terrestrial and aquatic species and do not reflect previously reported differences in limb length among cave and non-cave species.
Chapter 2: I reconstructed the phylogeographic history of Eurycea lucifuga across its range using tree-building methods as well as genetic clustering algorithms. Additionally,
I used molecular dating methods to infer lineage ages within the species. Results suggest that there are three major lineages within E. lucifuga, each of which show signs of either recent population expansion and/or persistent among-population gene flow.
Additionally, dating methods place the split among the major lineages on the scale of millions of years. Eurycea lucifuga has inhabited caves for as much time, or in some case much longer, than other taxa that show extreme forms of morphological adaptation to caves.
Chapter 3: I compared phenotypic differentiation with neutral genetic variation at a range-wide scale and within a single cave system in order to detect patterns of migration and signatures of selection on color in E. lucifuga. These analyses revealed that the spatial structure of differentiation varies among phenotypic traits and neutral genetic loci, suggesting that variation in color is influenced by selection. Higher relatedness between populations connected by above-ground routes rather than between those connected 8 within the cave system, in addition to the presentation of a river as an effective barrier to dispersal, suggest that a possible source of this selection may come from use of the surface habitat for dispersal.
Though many studies focus on the adaptive response to selection pressure, these are often focused on character change in response to one or a few variables in its primary environment. The most important message from this body of work is that it is crucial to keep a species’ evolutionary history and its complete ecology in mind when studying trait evolution. By incorporating phylogenetic relatedness into comparative models studying a broad group of taxa I showed that body shape evolution is more influenced by the transition between terrestrial to aquatic habitats than by colonization of caves, as previously thought. Similarly, I demonstrated that even limited use of an alternative habitat, in this case possibly as a conduit for dispersal, is enough to influence patterns of population structure in Eurycea lucifuga. Thus, it was only by uncovering the complicated ecology and evolutionary history of this species and other cave-dwelling salamanders that I was able to explain the disjunction between its habitat and morphology.
9 References
Agosta SJ, Klemens JA (2009) Resource specialization in a phytophagous insect: no
evidence for genetically based performance trade-offs across hosts in the field or
laboratory. Journal of Evolutionary Biology 22, 907-912.
Antonovics J (1976) The Nature of Limits to Natural Selection. Annals of the Missouri
Botanical Garden 63, 224-247.
Avise JC (2000) Phylogeography: the history and formation of species Harvard
University Press, Cambridge, MA
London, England.
Baker AJ, Peck MK, Goldsmith MA (1990) Genetic and morphometric differentiation in
introduced populations of Common Chaffinches (Fringilla coelebs) in New
Zealand. The Condor 92, 76-88.
Baldwin JM (1896) A new factor in evolution. The American Naturalist 30, 536-553.
Barr T, Holsinger J (1985) Speciation in cave faunas. Annual Review of Ecology and
Systematics 16, 313-337.
Bonett RM, Steffen MA, Lambert SM, Wiens JJ, Chippindale PT (2014) Evolution of
paedomorphosis in plethodontid salamanders: ecological correlates and re-
evaluation of metamorphosis. Evolution 68, 466-482.
Bradshaw A (1991) The Croonian Lecture, 1991: genostasis and the limits to evolution.
Philosophical Transactions of the Royal Society of London. Series B: Biological
Sciences 333, 289-305.
Buckling A, Rainey PB (2002) The role of parasites in sympatric and allopatric host
diversification. Nature 420, 496-499. 10 Caccone A (1985) Gene Flow in Cave Arthropods: A Qualitative and Quantitative
Approach. Evolution 39, 1223-1235.
Chakraborty R, Nei M (1977) Bottleneck Effects on Average Heterozygosity and Genetic
Distance with the Stepwise Mutation Model. Evolution 31, 347-356.
Charlesworth B, Hughes KA (2000) The maintenance of genetic variation in life-history
traits. Evolutionary genetics: from molecules to morphology 1, 369-392.
Clegg SM, Degnan SM, Kikkawa J, et al. (2002) Genetic Consequences of Sequential
Founder Events by an Island-Colonizing Bird. Proceedings of the National
Academy of Sciences of the United States of America 99, 8127-8132.
Cook LM, Mani GS, Varley ME (1986) Postindustrial Melanism in the Peppered Moth.
Science 231, 611-613.
Culver D, Master L, Christman M (2000) Obligate cave fauna of the 48 contiguous
United States. Conservation Biology.
Eggleston DB, Elis WE, Etherington LL, Dahlgren CP, Posey MH (1999) Organism
responses to habitat fragmentation and diversity: Habitat colonization by estuarine
macrofauna. Journal of Experimental Marine Biology and Ecology 236, 107-132.
Ehinger M, Mohr TJ, Starcevich JB, et al. (2014) Specialization-generalization trade-off
in a Bradyrhizobium symbiosis with wild legume hosts. BMC ecology 14, 8.
Ehrlich PR (1989) Attributes of invaders and the invading processes: vertebrates.
Biological invasions: a global perspective, 315-328.
Futuyma D (2008) Sympatric speciation: norm or exception. Specialization, Speciation,
and Radiation The Evolutionary Biology of Herbivorous Insects, 136-147. 11 Futuyma D (2010) Evolutionary constraint and ecological consequences. Evolution 64,
1865-1884.
Hawthorne DJ (1997) Ecological History and Evolution in a Novel Environment: Habitat
Heterogeneity and Insect Adaptation to a New Host Plant. Evolution 51, 153-162.
Holway DA, Suarez AV (1999) Animal behavior: an essential component of invasion
biology. Trends in Ecology & Evolution 14, 328-330.
Honnay O, Verheyen K, Butaye J, et al. (2002) Possible effects of habitat fragmentation
and climate change on the range of forest plant species. Ecology Letters 5, 525-
530.
Howarth F (1993) High-stress subterranean habitats and evolutionary change in cave-
inhabiting arthropods. The American Naturalist 142, S65-S77.
Howarth FG (1973) The cavernicolous fauna of Hawaiian lava tubes, 1. Introduction.
Pacific Insects 15, 139-151.
Huey RB, Hertz PE (1984) Is a Jack-of-All-Temperatures a Master of None? Evolution
38, 441-444.
Hutchison V (1958) The Distribution and Ecology of the Cave Salamander, Eurycea
lucifuga. Ecological Monographs 28, 1-20.
Jeffery WR (2005) Adaptive Evolution of Eye Degeneration in the Mexican Blind
Cavefish. Journal of Heredity 96, 185-196.
Jeffery WR (2009) Regressive Evolution in Astyanax Cavefish. Annual Review of
Genetics 43, 25-47.
Kettlewell HBD (1961) The Phenomenon of Industrial Melanism in Lepidoptera. Annual
Review of Entomology 6, 245-262. 12 Kolbe JJ, Glor RE, Rodriguez Schettino L, et al. (2004) Genetic variation increases
during biological invasion by a Cuban lizard. Nature 431, 177-181.
Kolbe JJ, Larson A, Losos JB (2007) Differential admixture shapes morphological
variation among invasive populations of the lizard Anolis sagrei. Molecular
Ecology 16, 1579-1591.
Lodge DM (1993) Biological invasions: Lessons for ecology. Trends in Ecology &
Evolution 8, 133-137.
Mitchell CE, Power AG (2003) Release of invasive plants from fungal and viral
pathogens. Nature 421, 625-627.
Nei M, Maruyama T, Chakraborty R (1975) The bottleneck effect and genetic variability
in populations. Evolution, 1-10.
Noriyuki S, Osawa N (2012) Intrinsic prey suitability in specialist and generalist
Harmonia ladybirds: a test of the trade-off hypothesis for food specialization.
Entomologia Experimentalis et Applicata 144, 279-285.
Petranka J (1998) Salamanders of the United States and Canada Smithsonian Institution
Press, Washington.
Porter M (2007) Subterranean biogeography: What have we learned from molecular
techniques? Journal of Cave and Karst Studies 69, 179-186.
Porter SS, Rice KJ (2013) Trade-offs, spatial heterogeneity, and the maintenance of
microbial diversity. Evolution 67, 599-608.
Protas M, Conrad M, Gross JB, Tabin C, Borowsky R (2007) Regressive Evolution in the
Mexican Cave Tetra, Astyanax mexicanus. Current Biology 17, 452-454. 13 Protas ME, Trontelj P, Patel NH (2011) Genetic basis of eye and pigment loss in the
cave crustacean, Asellus aquaticus. Proceedings of the National Academy of
Sciences 108, 5702-5707.
Reed DH, Lowe EH, Briscoe DA, Frankham R (2003) Fitness and adaptation in a novel
environment: effect of inbreeding, prior environment, and lineage. Evolution 57,
1822-1828.
Santos J, Pascual M, Simões P, et al. (2012) From nature to the laboratory: the impact of
founder effects on adaptation. Journal of Evolutionary Biology 25, 2607-2622.
Sket B (2008) Can we agree on an ecological classification of subterranean animals?
Journal of Natural History 42, 1549-1563.
Smith JM, Burian R, Kauffman S, et al. (1985) Developmental constraints and evolution:
a perspective from the Mountain Lake conference on development and evolution.
The Quarterly Review of Biology 60, 265-287.
Soulé M (1971) The variation problem: the gene flow-variation hypothesis. Taxon, 37-50.
Thomas CD, Bodsworth EJ, Wilson RJ, et al. (2001) Ecological and evolutionary
processes at expanding range margins. Nature 411, 577-581.
Torchin ME, Lafferty KD, Dobson AP, McKenzie VJ, Kuris AM (2003) Introduced
species and their missing parasites. Nature 421, 628-630.
Travisano M, Vasi F, Lenski RE (1995) Long-Term Experimental Evolution in
Escherichia coli. III. Variation Among Replicate Populations in Correlated
Responses to Novel Environments. Evolution 49, 189-200.
Tripp EA, Manos PS (2008) Is floral specialization an evolutionary dead-end? pollination
system transitions in Ruellia (Acanthaceae). Evolution 62, 1712-1737. 14 Wessel A, Erbe P, Hoch H (2007) Pattern and process: Evolution of troglomorphy in
the cave-planthoppers of Australia and Hawai'i–Preliminary observations
(Insecta: Hemiptera: Fulgoromorpha: Cixiidae). Acta Carsologica 36, 199-206.
Yamamoto Y, Jeffery W (2000) Central Role for the Lens in Cave Fish Eye
Degeneration. Science (New York, N.Y.) 289, 631-633.
Yamamoto Y, Stock DW, Jeffery WR (2004) Hedgehog signalling controls eye
degeneration in blind cavefish. Nature 431, 844-847.
15
CHAPTER 1
THE PHYLOGENETIC HISTORY OF BODY SHAPE VARIATION IN
SALAMANDERS OF THE GENUS EURYCEA (PLETHODONTIDAE)
Abstract
Body shape is directly related to many aspects of a species’ evolution and ecology. Body and limb elongation are associated with a general phenotype exhibited by species inhabiting cave environments. However, studies explicitly testing for differences in body shape between cave-dwelling and non-cave-dwelling lineages are rare. Here we examine the variation in body shape among 20 species in the salamander genus Eurycea
(Plethodontidae) in species inhabiting aquatic or terrestrial, and cave or non-cave habitats. After analyzing morphometric data in a phylogenetic context, we found that there is no evidence of differences in limb elongation between cave and non-cave species.
Instead, we find significant differences in general body size between aquatic and terrestrial species, and significant differences in tail length between cave and non-cave species. These results suggest that habitat does impact body shape in this group of salamanders, but perhaps differently than previously thought.
16 Background
Body shape is a key part of morphological variation among vertebrates, with impacts on function and ecology (Carroll 1997; Collar et al. 2013). Variation in shape may be a result of environmental effects, structural or functional constraints, adaptive differentiation, or shared phylogenetic history (Blomberg et al. 2003; Gould 2002; Losos
2011). There are many examples of body shape divergence that have been attributed completely to adaptation to ecological circumstance (Schluter et al. 2004; Shine 1986;
Walker & Bell 2000), differences of function (e.g. the use of limbs for running across open ground versus clinging to rocky outcrops) (Cunha et al. 2009; Kamiya 2011; Losos
1990; Walker 1997), or a combination of the two (Klingenberg et al. 2003;
Wikramanayake 1990). Often, patterns of morphological variation are shaped by shared phylogenetic history (Álvarez et al. 2013; Stayton 2005), which may influence variation in function or behavior (Bergmann & Irschick 2010). Understanding what causes variation in body shape is important for understanding how it may impact a species’ evolutionary trajectory: for example, increased fitness from the evolution of a certain body shape may prevent divergence from that shape, whereas similarity due to shared evolutionary history may not limit future changes in morphology.
Elongation of body or limb is a specific axis of morphological variation that has long been included in a suite of traits associated with cave-dwelling species (Brandon
1971; Christiansen 1961; Mitchell & Reddell 1965; Sket 2008; Weber 2004; White &
Culver 2012; Wilkens et al. 2000). Cave-dwelling taxa have been of particular interest for many years because of their dramatic morphological and physiological changes, the simplicity of the selection regime within the cave habitat and the resulting parallel 17 evolution of cave-associated traits (Barr & Holsinger 1985; Culver 1982). These traits, known collectively as troglomorphy, include other features such as regression of eyes, depigmentation, enhanced extra-optic sensory structures, and reduced metabolism.
Troglomorphic traits result from both a relaxation of selection pressures formerly experienced in the ancestral surface habitats, and as a result of directional selection experienced within the cave environment (Pipan & Culver 2012). Though most cases of troglomorphic elongation have been studied in invertebrates (Barr & Holsinger 1985;
Christiansen 1961), studies of cave vertebrates, and salamanders in particular, also associate elongation with cave-dwelling (Bendik et al. 2013; Mitchell & Reddell 1965;
Wiens et al. 2003).
In this study we compared body shape variation among different habitat groups by comparing morphological measurements among (1) aquatic and terrestrial and (2) cave and surface-dwelling species in the genus Eurycea and several outgroups. This work addresses a number of issues with our current knowledge of the evolution of body elongation as it relates to habitat occupancy: First, we analyzed the relationship between habitat and trait evolution using phylogenetically corrected models. Though it is important to consider trait evolution in the context of patterns of relatedness in order to avoid bias (Brooks et al. 1995; Felsenstein 1985), to our knowledge troglomorphic elongation has not been assessed using phylogenetically based statistical methods.
Additionally, the majority of the previous studies comparing cave-dwelling and non- cave-dwelling salamanders have focused on aquatic cave-restricted species exhibiting all troglomorphic traits. By studying a broader ecological and morphological sampling of the genus, with appropriate outgroups, we were able to compare species in a variety of 18 habitats, providing a greater insight into the relationship between ecology, phylogeny, and morphology.
Methods
Morphometric data collection
In December, 2013 and February, 2015 we took photographs of 480 preserved specimens representing 20 species of Eurycea and 8 representative species of outgroup genera in the family Plethodontidae in the herpetology collections of the American
Museum of Natural History (New York City, New York) and the Smithsonian Institution
National Museum of Natural History (Washington, D.C.) (Supplementary Table S1).
Photographs included three angles (dorsal, ventral, and lateral views), and included a penny as a size standard. Because sexual size dimorphism is minor relative to individual size variance in salamanders (Petranka 1998) we did not attempt to collect data on sex from these specimens. We measured nine morphometric traits from these photographs using the image processing software ImageJ (NIH). These traits include: head width, front leg length, front leg width, body width at the widest part between the front and back legs, back leg length, back leg width, the length of the fourth back digit, tail length, and snout-vent length. One person (J. Allen) performed all of the digital processing to avoid among-researcher error in measurement. Using information from the IUCN Redlist website (IUCN Global Species Programme Red List Unit), AmphibiaWeb, and Petranka
(1998), we assigned each species to a ‘habitat’ category, including whether it is aquatic or non-aquatic, and cave-dwelling or non-cave-dwelling (Supplementary Table S1). These designations reflect the habitat in which each species is most likely to spend the majority of its adult life stage. 19 Phylogenetic reconstruction
In order to analyze these data in a phylogenetic context we aggregated sequence data from the gene fragments 16s, cytb, POMC, and rag1 for each species from GenBank
(Benson et al. 2011). Sequence information and accession numbers are available in
Supplementary Table 1. We aligned, trimmed, and edited each locus using MEGAv.5.2.2
(Tamura et al. 2011). Species with no data available for a locus were represented with a string of missing data characters so as to make all four loci identical in number and order of species. We used jModelTest v2.1.4 (Posada 2008) using the Akaike Information
Criterion, Bayesian Information Criterion, and Decision Theory methods to find the most appropriate models of substitution for each locus, then reconstructed the phylogenetic history using each locus as a separate partition with raxmlGUI v1.31 (Silvestro &
Michalak 2012). While it was not included in the comparative analyses, we included data from Proteus anguinus in the phylogenetic reconstruction as an outgroup in order to root the tree. We ran the analysis with 1000 bootstrap replicates, and visualized the resulting
Maximum Likelihood tree using FigTree v1.4.0 (Ronquist et al. 2012).
Uncorrected statistical analyses
All statistical analyses were performed using R v.3.1.2 (R Core Team, 2014) interfaced through RStudio v.0.98.1091 (RStudio, Inc.). We first collapsed the data set to include trait means for each species using the function summaryBy() in the package doBy
(Hojsgaard & Halekoh 2015) in order to mitigate any effects of unequal sampling within the species. We performed Principal Components Analysis (PCA) on the nine log- transformed body shape measurements in order to obtain axes of variation using the function prcomp() in the R package stats (R Core Team, 2014). We tested these data for 20 normality and heteroscedasticity using the shapiro.test() function in the R package stats and leveneTest() function in the R package car (Fox & Weisberg 2011), respectively. We then employed two separate linear models to assess the relationship between habitat and each principal component, including PC1 and PC2 as the dependent variable in each model and Aquatic/Terrestrial, Cave/Non-cave, and the interaction term as the independent variables using the function lm() in the R package stats. We performed Type III tests to assess the significance of the linear model using the function
Anova() in the package car, and visualized plots of these data with ggplot() in the package ggplot2 (Wickham 2009).
Phylogenetically-corrected statistical analysis
Since relatedness among species may impinge on the independence of these data, we also analyzed them in a phylogenetic context. We performed phylogenetic comparative analyses using the Maximum Likelihood tree obtained previously with four separate analyses. In each model we included either PC1 PC2 as each dependent variable.
Since phylogenetic ANOVA cannot include multiple independent factors, we ran two models for each dependent variable, one comparing Cave/Non-cave groups, and one comparing Aquatic/Terrestrial groups. These analyses used simulations to obtain a distribution of empirical F statistics with which to test each hypothesis. We chose to run each analysis using the default assumption of evolution by Brownian motion. We used the function aov.phylo() in the R package geiger (Garland et al. 1993), specifying the
Wilks test, and performed 1000 simulations. 21 Results
Morphology
Principal Component Analysis (PCA) was performed on all 25 species, including 20 species of Eurycea and 5 outgroup species. The first two principal components accounted for almost 100% of the cumulative variance (Table 1). PC1, which accounted for 83% of the variance, was representative largely of increased snout-vent length and tail length, and had positive loadings of all other traits, so we consider this to be representative of general size. PC2, which accounted for 16% of the variance, had a strong positive loading of tail length, and a strong negative loading of snout-vent length.
Summary plots of each principal component can be seen in Figure 1a,b. Tests of normality indicated that PC2 violates this assumption (W=0.906, p=0.025), but visual inspection of the data indicated that these violations are mild. Additionally, Levene tests indicated that our data do not violate assumptions of homoscedasticity.
Phylogenetic models
Sequence data were collected from Genbank for all 25 species for Rag1 and cytb, and for the majority of species for POMC and 16s. jModelTest predicted the most appropriate model of evolution for each locus to be GTR+I+G (POMC, Rag1) and
GTR+G (cytb, 16s). The Maximum Likelihood tree generated from the four loci recapitulated what most phylogenetic reconstructions of this group have shown, with bootstrap values of over 50 for almost all branches, and most near 100 (Figure 2). We estimate by visual inspection that there are four to six transitions to cave-dwelling within the species we sampled, and possibly three to five transitions from terrestrial to aquatic.
22 Comparative statistical analyses
ANOVA comparing both PC1 and PC2 among the different habitat groups indicated that terrestrial species had significantly higher values of PC1 (general body size) than aquatic species. However, there were no significant differences among the habitat groups in PC2 (tail length relative to body size; Table 2, Figure 3). Analyzing these data using phylogenetic comparative analyses, we found that again PC1 (general body size) was significantly larger in terrestrial species than aquatic species. We also found that when analyzed in the phylogenetic ANOVA cave species had significantly higher values of PC2 (tail length relative to body length) than non-cave species, meaning that they have relatively shorter tails than non-cave species (Table 3).
Discussion
We compared body shape measurements among 20 species of Eurycea salamanders, together with 8 outgroup species from distinct genera. Our results indicate that the first principal component, which represents body size generally, accounts for the vast majority of the morphological variation that we measured among species. Relative tail length is represented by the second principal component, which accounts for 16% of the variation. One of our aims was to examine the differences between cave and non- cave species; however, morphological variation classically associated with cave-dwelling
(relative length of limbs) was only included in a principal component accounting for one percent of the total variance. The primary differences we found in morphology among habitat groups were that terrestrial species were generally larger than aquatic species, and that cave species had shorter tails than non-cave species. Visual inspection suggests that this difference in tail length is comprised largely of differences between terrestrial cave 23 and non-cave species. While the difference in general size between aquatic and terrestrial species was revealed by both the phylogenetically uncorrected ANOVA and the phylogenetic comparative analyses, only in the phylogenetic ANOVA did we find a difference in tail length between cave and non-cave species. The discordance between the ANOVA and phylogenetic ANOVA results suggests that phylogenetic signal was masking the effect of tail length within this group.
Variation in morphology arises through many different mechanisms including environmental influences, structural or functional constraints, or shared evolutionary history (Blomberg et al. 2003; Gould 2002; Losos 2011). Some taxa show distinct differences among populations due primarily to ecological differences (Álvarez et al.
2013; Clabaut et al. 2007; Kamiya 2011) which may be driven primarily by functional differences in how traits benefit organisms in those habitats (e.g., climbing requires different attributes than swimming or burrowing) (Blankers et al. 2012). Other taxa exhibit a combination of ecology-driven and phylogeny-driven variation among lineages
(Blankers et al. 2012; Jockusch 1997). The similarity between the results of the phylogenetically uncorrected ANOVA and the phylogenetic comparative models we performed suggests that any significant differences in morphology cannot be explained only by shared evolutionary history. A role of ecological difference in variation of tail length particularly may be due to functional differences in tails between aquatic and terrestrial salamanders. While in aquatic species the tail is mainly used for locomotion, in terrestrial species it also has functional significance in fat storage and predator defense strategies (Arnold 1982; Brodie 1977; Maiorana 1977; Vaglia et al. 1997). Such 24 differences in function may necessitate differences in form, such as the differences in length we see among non-cave terrestrial and aquatic species.
A surprising part of our results is the small amount of body shape variation represented by limb length relative to body size, which has been described as one of the traits distinguishing cave-dwelling species (Brandon 1971; Christiansen 1961; Mitchell &
Reddell 1965; Sket 2008; Weber 2004; White & Culver 2012; Wilkens et al. 2000). Past comparisons that find differences in shape among cave-dwelling and non-cave-dwelling populations (Bendik et al. 2013; Mitchell & Reddell 1965; Wiens et al. 2003), have focused mainly on the Texas clade of cave dwelling and non-cave dwelling Eurycea which are entirely aquatic, and none have included phylogenetically corrected statistical models. Our study, in contrast, suggests that variation between cave and non-cave species exists in the length of the tail, rather than limb elongation.
Historical transitions between habitats, characterized by Bonett et al. (2014), are complicated in the Eurycea. The ancestral group to the Spelerpines, which includes
Eurycea and Gyrinophilus, was most likely a non-cave terrestrial species. Within the
Spelerpines, the change from a terrestrial to an aquatic habitat occurred multiple times followed by at least one reversal back to terrestriality. The transition to cave-dwelling also occurred multiple times, mainly within those groups that transitioned to paedomorphosis, and possibly followed by at least one reversal to non-cave occupancy.
Though there was no statistical support for either the transition from terrestriality to aquatic or from non-cave to cave-dwelling being earlier, and the two were likely temporally close, the authors predicted that the transition to cave-dwelling was the more recent change, and that paedomorphosis had allowed populations to make use of limited 25 cave resources (Bonett et al. 2014). An initial transition to the aquatic habitat followed by cave colonization is also thought to have occurred in the non-Spelerpine cave salamander Proteus anguinus (Proteidae), as all proteids are paedomorphic, and most Proteus are cave-dwellers (Sket & Arntzen 1994).
The complex transition history within these salamanders has interesting implications for the results presented here because of the patterns we see among habitat groups in body shape. Because there are similar body shapes among cave and non-cave species within the aquatic and terrestrial groups, it is likely that convergent evolution of shape traits followed each habitat transition. If, as Bonett et al. (2014) hypothesize, the first transition was of terrestrial species into aquatic habitats followed by independent cave colonizations, then aquatic species first became smaller generally than terrestrial species. Subsequently, the colonization of caves by aquatic species led to shortened tails, and cave colonization by terrestrial species resulted in both shortened tails and larger general size. Alternatively, if the transition to the cave habitat was followed by independent transitions to aquatic habitats, then cave-dwelling species initially experienced a reduction in tail length. This would have been followed by a reduction in general size among aquatic species both in and out of caves. Either scenario makes it likely that in each transition there were multiple independent changes in body form, which suggests that body shape in these species is strongly influenced by ecological factors. Similar independent evolution due to ecology has been found in other systems, such as the convergent reductions in bone size in freshwater threespine sticklebacks
(Walker & Bell 2000) and repeated elongation within families of reef fish (Claverie &
Wainwright 2014). 26 Conclusions
This study examined variation in body shape among 25 species of salamander inhabiting two habitat categories: cave vs. non-cave and aquatic vs. terrestrial in order to characterize whether ecological habitat was predictive of form across a broad sampling of taxa. Although limb elongation in vertebrates has previously been considered a trait included in a suite of cave-associated traits, an examination of a broader sampling of
Eurycea including both aquatic and terrestrial species as well as cave and non-cave species indicates that limb length does not contribute substantially to variation among habitat groups. There were significant differences among habitat groups in both general body size and tail length, wherein terrestrial species were larger than aquatic species, and non-cave species had larger tails than cave species. These results suggest that explicit testing is important for our understanding of the relationships between body shape and habitat in salamander species.
Acknowledgements
The authors would like to acknowledge the contributions of the staff at the
American Museum of Natural History, New York NY, including but not limited to David
Kizirian and David Dickey. Julia Allen performed all of the digital image processing.
The Brodie and Cox labs at the University of Virginia provided valuable feedback on the analyses and presentation of these data.
27
References
AmphibiaWeb: Information on amphibian biology and conservation [web application].
AmphibiaWeb, Berkeley, California. http://amphibiaweb.org
Álvarez A, Perez SI, Verzi DH (2013) Ecological and phylogenetic dimensions of
cranial shape diversification in South American caviomorph rodents
(Rodentia: Hystricomorpha). Biological Journal of the Linnean Society 110,
898-913.
Arnold SJ (1982) A quantitative approach to antipredator performance: salamander
defense against snake attack. Copeia, 247-253.
Barr T, Holsinger J (1985) Speciation in cave faunas. Annual Review of Ecology and
Systematics 16, 313-337.
Bendik NF, Meik JM, Gluesenkamp AG, Roelke CE, Chippindale PT (2013)
Biogeography, phylogeny, and morphological evolution in central Texas cave
and spring salamanders. BMC Evol Biol 13, 201.
Benson DA, Karsch-Mizrachi I, Clark K, et al. (2011) GenBank. Nucleic Acids
Research.
Bergmann PJ, Irschick DJ (2010) Alternate pathways of body shape evolution
translate into common patterns of locomotor evolution in two clades of
lizards. Evolution 64, 1569-1582.
Blankers T, Adams DC, Wiens JJ (2012) Ecological radiation with limited
morphological diversification in salamanders. Journal of Evolutionary Biology
25, 634-646. 28 Blomberg SP, Garland T, Ives AR (2003) Testing for phylogenetic signal in
comparative data: behavioral traits are more labile. Evolution 57, 717-745.
Bonett RM, Steffen MA, Lambert SM, Wiens JJ, Chippindale PT (2014) Evolution of
paedomorphosis in plethodontid salamanders: ecological correlates and re-
evaluation of metamorphosis. Evolution 68, 466-482.
Brandon RA (1971) North American troglobitic salamanders: some aspects of
modification in cave habitats, with special reference to Gyrinophilus
palleucus. Bulletin of the National Speleological Society 33, 1-21.
Brodie ED, Jr. (1977) Salamander Antipredator Postures. Copeia 1977, 523-535.
Brooks DR, McLennan DA, Carpenter JM, Weller SG, Coddington JA (1995)
Systematics, Ecology, and Behavior. BioScience 45, 687-695.
Carroll RL (1997) Patterns and Processes of Vertebrate Evolution Cambridge
University Press.
Christiansen K (1961) Convergence and parallelism in cave Entomobryinae.
Evolution.
Clabaut C, Bunje PME, Salzburger W, Meyer A (2007) Geometric morphometric
analyses provide evidence for the adaptive character of the Tanganyikan
cichlid fish radiations. Evolution 61, 560-578.
Claverie T, Wainwright PC (2014) A Morphospace for Reef Fishes: Elongation Is the
Dominant Axis of Body Shape Evolution. PLoS ONE 9, e112732.
Collar DC, Reynaga CM, Ward AB, Mehta RS (2013) A revised metric for quantifying
body shape in vertebrates. Zoology (Jena) 116, 246-257. 29 Culver D (1982) Cave life: Evolution and Ecology Harvard University Press,
Cambridge, MA.
Cunha C, Bastir M, Coelho MM, Doadrio I (2009) Body shape evolution among ploidy
levels of the Squalius alburnoides hybrid complex (Teleostei, Cyprinidae).
Journal of Evolutionary Biology 22, 718-728.
Felsenstein J (1985) Phylogenies and the Comparative Method. The American
Naturalist 125, 1-15.
Fox J, Weisberg S (2011) An {R} Companion to Applied Regression, Second Edition
Sage, Thousand Oaks, CA.
Garland TJ, Dickerman A, Janis C, Jones J (1993) Phylogenetic analysis of covariance
by computer simulation. Systematic Biology 42, 265-292.
Gould SJ (2002) The structure of evolutionary theory Harvard University Press.
Hojsgaard S, Halekoh U (2015) Groupwise statistics, LSmeans, linear contrasts,
utilities v.4.5-13.
Jockusch EL (1997) Geographic Variation and Phenotypic Plasticity of Number of
Trunk Vertebrae in Slender Salamanders, Batrachoseps (Caudata:
Plethodontidae). Evolution 51, 1966-1982.
Kamiya T (2011) Morphological and Ethological Adaptations of Ostracoda to
Microhabitats in Zostera Beds. Evolutionary Biology of Ostracoda: Its
Fundamentals and Applications, 303.
Klingenberg CP, Barluenga M, Meyer A (2003) Body shape variation in cichlid fishes
of the Amphilophus citrinellus species complex. Biological Journal of the
Linnean Society 80, 397-408. 30 Losos JB (1990) The Evolution of Form and Function: Morphology and Locomotor
Performance in West Indian Anolis Lizards. Evolution 44, 1189-1203.
Losos JB (2011) Seeing the Forest for the Trees: The Limitations of Phylogenies in
Comparative Biology. The American Naturalist 177, 709-727.
Maiorana VC (1977) Tail autotomy, functional conflicts and their resolution by a
salamander. Nature 265, 533-535.
Mitchell RW, Reddell JR (1965) Eurycea tridentifera, a new species of troglobitic
salamander from Texas and a reclassification of Typhlomolge. Texas journal
of science, 12-27.
Petranka J (1998) Salamanders of the United States and Canada Smithsonian
Institution Press, Washington.
Pipan T, Culver DC (2012) Convergence and divergence in the subterranean realm: a
reassessment. Biological Journal of the Linnean Society.
Posada D (2008) jModelTest: phylogenetic model averaging. Molecular Biology and
Evolution 25, 1253-1256.
Ronquist F, Teslenko M, Van Der Mark P, et al. (2012) MrBayes 3.2: Efficient
bayesian phylogenetic inference and model choice across a large model
spece. Systematic Biology.
Schluter D, Clifford E, Nemethy M, McKinnon JS (2004) Parallel Evolution and
Inheritance of Quantitative Traits. The American Naturalist 163, 809-822.
Shine R (1986) Sexual Differences in Morphology and Niche Utilization in an Aquatic
Snake, Acrochordus arafurae. Oecologia 69, 260-267. 31 Silvestro D, Michalak I (2012) raxmlGUI: a graphical front-end for RAxML.
Organisms Diversity & Evolution 12, 335-337.
Sket B (2008) Can we agree on an ecological classification of subterranean animals?
Journal of Natural History 42, 1549-1563.
Sket B, Arntzen J (1994) Proteus anguinus parkelj n. ssp.(Urodela: Proteidae).
Bijdragen tot de Dierkunde 64, 33-53.
Stayton CT (2005) Morphological evolution of the lizard skull: A geometric
morphometrics survey. Journal of Morphology 263, 47-59.
Tamura K, Peterson D, Peterson N, et al. (2011) MEGA5: Molecular Evolutionary
Genetics Analysis Using Maximum Likelihood, Evolutionary Distance, and
Maximum Parsimony Methods. Molecular Biology and Evolution 28, 2731-
2739.
Vaglia JL, Babcock SK, Harris RN (1997) Tail development and regeneration
throughout the life cycle of the four-toed salamander Hemidactylium
scutatum. Journal of Morphology 233, 15-29.
Walker JA (1997) Ecological morphology of lacustrine threespine stickleback
Gasterosteus aculeatus L. (Gasterosteidae) body shape. Biological Journal of
the Linnean Society 61, 3-50.
Walker JA, Bell MA (2000) Net evolutionary trajectories of body shape evolution
within a microgeographic radiation of threespine sticklebacks (Gasterosteus
aculeatus). Journal of Zoology 252, 293-302.
Weber A (2004) Amphibia. In: Encyclopedia of caves and karst science (ed. Gunn J),
pp. 125-128. 32 White WB, Culver DC (2012) Encyclopedia of Caves Academic Press.
Wickham H (2009) ggplot2: elegant graphics for data analysis Springer, New York.
Wiens JJ, Chippindale PT, Hillis DM (2003) When are phylogenetic analyses misled
by convergence? a case study in Texas cave salamanders. Systematic Biology
52, 501-514.
Wikramanayake ED (1990) Ecomorphology and Biogeography of a Tropical Stream
Fish Assemblage: Evolution of Assemblage Structure. Ecology 71, 1756-1764.
Wilkens H, Culver DC, Humphreys WF (2000) Subterranean ecosystems Elsevier.
33 Figures
40 Cave%aqua-c% Cave%terrestrial% Non2cave%aqua-c% 20 Non2cave%terrestrial%habitat
Cave_Aquatic
Cave_Terrestrial 0 PC1 Non−cave_Aquatic
Non−cave_Terrestrial
−20
20
habitat Eurycea_nana Eurycea_latitans Cave_Aquatic Eurycea_wallaceiEurycea_lucifugaEurycea_spelaea 10Eurycea_rathbuni Eurycea_aquaticaEurycea_cirrigera Eurycea_wilderae Eurycea_pterophila Eurycea_bislineataEurycea_neotenesEurycea_junaluska Eurycea_tridentifera Hydromantes_genei Eurycea_tynerensis Eurycea_multiplicata Cave_Terrestrial Hydromantes_italicus Eurycea_guttolineata Hydromantes_brunus Eurycea_l._longicauda PC2 Eurycea_quadridigitata Non−cave_Aquatic Eurycea_l._melanopleura Gyrinophilus_porphyriticus 0 Hydromantes_platycephalus Non−cave_Terrestrial
−10
Eurycea_nana
Eurycea_latitans Eurycea_lucifuga Eurycea_rathbuniEurycea_wallaceiEurycea_spelaeaEurycea_aquaticaEurycea_cirrigera Eurycea_wilderae Eurycea_pterophila Eurycea_bislineataEurycea_neotenesEurycea_junaluska Eurycea_tridentifera Hydromantes_genei Eurycea_tynerensis Eurycea_multiplicata Hydromantes_italicus Eurycea_guttolineata Hydromantes_brunus Eurycea_l._longicaudaEurycea_quadridigitata Eurycea_l._melanopleura Gyrinophilus_porphyriticus Hydromantes_platycephalus
Figure 1. Species means of PC1 (general body size) and PC2 (relative tail length), grouped according to habitat preference. Habitat designation is indicated by color, as seen in the key.
34
40 a a b b a b a b
Cave% Non)cave% 20
20
10
PC1 0 PC2
0
−20
−10
Aquatic Terrestrial Aquatic Terrestrial
Figure 2. The relationships between PC1 (general body size) and PC2 (relative tail length) and habitat among species means were explored using a linear model and then in a phylogenetically corrected
ANOVA. Box and whisker plots indicate the median, first and third quartiles, and whiskers extend to 1.5 for the interquartile range. We found significant differences among aquatic and terrestrial species in PC1
(p=0.015), as well as significant differences among cave and non-cave species in PC2 (p=0.011). Proteus_anguinus
100 Necturus_punctatus Necturus_beyeri 35 Hydromantes_genei 99 Cave%aqua(c%
100 Hydromantes_italicus Cave%terrestrial%
100 Hydromantes_brunus Non2cave%aqua(c%
Hydromantes_platycephalus Non2cave%terrestrial%
100 Gyrinophilus_porphyriticus
Eurycea_multiplicata 78
89 Eurycea_tynerensis 100 Eurycea_spelaea
Eurycea_quadridigitata
98 Eurycea_rathbuni 100 100 Eurycea_tridentifera 80 Eurycea_neotenes 53 Eurycea_latitans 44 Eurycea_pterophila 88 16 Eurycea_nana
Eurycea_lucifuga 88 Eurycea_l._melanopleura 53
56 Eurycea_guttolineata
92 Eurycea_l._longicauda
Eurycea_wallacei
43 100 Eurycea_aquatica
Eurycea_junaluska 98 Eurycea_cirrigera 78
55 Eurycea_bislineata
Eurycea_wilderae
0.2 Figure 3. Phylogenetic relationships among these species were compared using a maximum likelihood tree-
building approach, with Proteus anguinus included as an outgroup (not shown). Branches are labeled with
bootstrap scores, and color represents habitat assignment (black=cave aquatic, orange=cave terrestrial,
blue=non-cave aquatic, green=non-cave terrestrial).
36 Tables Table 1. Components of variance of the Principal Components Analysis, and loadings of each trait on each
PC. PC1 and PC2 were used in further analyses.
PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 PC9
Standard 20.288 8.790 2.793 0.666 0.486 0.292 0.192 0.103 0.056 Proportion 0.827 0.155 0.016 0.001 0.000 0.000 0.000 0.000 0.000 Cumulative 0.827 0.983 0.998 0.999 1.000 1.000 1.000 1.000 1.000
PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 PC9
Head Width 0.079 0.179 -0.200 0.704 -0.443 -0.305 0.238 -0.272 -0.076 Front Limb Length 0.115 0.152 -0.669 -0.109 -0.393 0.460 -0.329 0.170 0.026 Front Limb Width 0.026 0.032 0.001 0.094 0.140 0.376 0.193 -0.424 0.782 Body Width 0.088 0.137 0.008 0.603 0.517 0.063 -0.483 0.312 0.084 Back Limb Length 0.141 0.158 -0.635 -0.214 0.511 -0.430 0.222 -0.101 0.021 Back Limb Width 0.033 0.030 -0.025 0.088 0.296 0.489 0.050 -0.541 -0.607 Digit Length 0.018 0.018 -0.047 0.163 0.101 0.354 0.716 0.562 -0.081 Tail Length 0.808 -0.587 0.007 0.038 -0.026 -0.011 -0.002 -0.003 0.002 Snout-Vent Length 0.545 0.745 0.325 -0.194 -0.058 0.006 0.022 0.018 -0.013
Table 2. ANOVA results from models including each principal component as the dependent variable, and both aquatic/terrestrial and cave/non-cave comparisons as independent variables, as well as an interaction term. Significance is denoted by bold.
SS DF F p
PC1
Cave/Non-cave 7 1,21 0.026 0.873 Aquatic/Terrestrial 2683.4 1,21 10.083 0.005 Interaction 241.1 1,21 0.906 0.352 PC2
Cave/Non-cave 37.17 1,21 0.655 0.428 Aquatic/Terrestrial 160.65 1,21 2.829 0.107 Interaction 166.13 1,21 2.926 0.102
37 Table 3. Results from separate Phylogenetic ANOVAs including each PC as the dependent variable, and habitat as the independent variable.
DF SS F p
PC1
Aquatic/Terrestrial 1,23 3900.8 15.009 0.015 Cave/Non-cave 1,23 102.3 0.241 0.671 PC2
Aquatic/Terrestrial 1,23 12.58 0.157 0.850 Cave/Non-cave 1,23 477.43 7.976 0.011
38 Supplementary Materials
Supplementary Table 1. Accession numbers of taxa included in phylogenetically corrected comparisons. Species 16s cytb POMC Rag1 Eurycea aquatica KF562543.1 FJ750235
Eurycea bislineata JQ920581.1 AY528402 EU275815 EU275784 Eurycea cirrigera JQ920583.1 KF562548.1 JQ920728.1 FJ750245 Eurycea guttolineata JQ920586.1 FJ866207.1 JQ920770.1
Eurycea junaluska KF562550.1 FJ750246
Eurycea latitans KF562551.1 KF562652.1
Eurycea longicauda longicauda JQ920586.1 AY528403 JQ920730.1 AY650121 Eurycea longicauda melanopleura FJ866478.1 KF562552.1 KF562653.1
Eurycea lucifuga JQ920584.1 EF044248 JQ920729.1 FJ917632 Eurycea multiplicata JQ920580.1 AY528341 JQ920725.1 AY691707 Eurycea nana JQ920590.1 AY014846 JQ920735.1 EF443113 Eurycea neotenes AY528400 AY650122
Eurycea pterophila AY014851 KF562658.1
Eurycea quadridigitata AY523777 AY528401 JQ920759.1 AY691708 Eurycea rathbuni AY014845 KF562663.1
Eurycea spelaea FJ866336.1 KF562667.1
Eurycea tridentifera KF562565.1 KF562669.1
Eurycea tynerensis AY528373 JF768990.1 KF562676.1
Eurycea wallacei AF252380 KF562693.1
Eurycea wilderae JQ920582.1 AF252379 JQ920727.1 KF562680.1 Gyrinophilus porphyriticus JQ920577.1 JQ920616.1 EU275853 AY583349 Hydromantes brunus HM989444.1 U89614.1 EU275825.1 HM797660.1 Hydromantes genei FJ602156.1 U89617.1 EU275840.1 FJ602343.1 Hydromantes italicus EF107189.1 FJ602303.1 EU275827.1 EF107312.1 Hydromantes platycephalus EF107215.1 U89612 EU275828 EU275793 Proteus anguinus EF107180 GQ368659 KC295576.1 AY650138
39
Supplementary Table 2. Raw measurement data and habitat assignments for each species included in the comparative analyses. Forelimb Hindlimb
Institution Catalog # Species Head Width Length Width Body Width Length Width Digit Length Tail Length Snout-Vent Length Eye Width Habitat
AMNH 69032 Eurycea aquatica 6.596 7.664 1.343 6.368 9.659 2 0.857 42.676 36.807 2.221 Non-cave_Aquatic
AMNH 69033 Eurycea aquatica 6.525 8.656 1.302 6.521 9.682 1.802 0.46 44.696 41.854 2.086 Non-cave_Aquatic
AMNH 90687 Eurycea aquatica 5.454 6.337 1.37 4.639 9.036 1.435 0.648 36.587 36.99 Non-cave_Aquatic
AMNH 108781 Eurycea aquatica 6.659 7.704 1.21 6.024 8.715 1.837 1.205 41.699 42.159 1.683 Non-cave_Aquatic
AMNH 108782 Eurycea aquatica 5.803 7.809 1.201 5.589 9.619 1.761 0.999 30.493 40.378 1.748 Non-cave_Aquatic
AMNH 108783 Eurycea aquatica 6.017 8.232 1.186 6.385 9.325 1.865 0.99 36.24 40.081 1.574 Non-cave_Aquatic
AMNH 108784 Eurycea aquatica 5.556 4.624 0.937 5.587 5.293 0.917 0.724 25.907 32.118 1.074 Non-cave_Aquatic
AMNH 108785 Eurycea aquatica 4.846 6.908 1.013 4.809 8.244 1.463 0.663 22.73 33.252 1.197 Non-cave_Aquatic
AMNH 108786 Eurycea aquatica 6.361 6.786 1.069 5.828 6.739 1.503 0.846 25.766 32.603 1.28 Non-cave_Aquatic
AMNH 108787 Eurycea aquatica 4.841 4.999 0.629 3.909 4.234 0.907 0.656 19.91 24.657 1.296 Non-cave_Aquatic
AMNH 108788 Eurycea aquatica 5.048 6.829 1.01 4.421 9.008 1.145 1.102 21.35 30.818 1.55 Non-cave_Aquatic
AMNH 6632 Eurycea bislineata 5.862 5.518 0.873 5.103 8.95 1.626 1.483 23.406 42.618 1.868 Non-cave_Aquatic
AMNH 7540 Eurycea bislineata 5.8 5.918 1.156 5.096 8.055 1.283 1.074 22.317 38.189 1.844 Non-cave_Aquatic
AMNH 7543 Eurycea bislineata 5.269 5.212 1.281 4.961 6.981 1.376 0.813 35.74 32.817 1.503 Non-cave_Aquatic
AMNH 13099 Eurycea bislineata 4.8 5.516 1.121 5.363 7.972 1.855 1.187 45.4 41.929 1.227 Non-cave_Aquatic
AMNH 13100 Eurycea bislineata 6.404 4.529 1.407 6.271 5.194 1.331 1.058 44.798 47.714 1.637 Non-cave_Aquatic
AMNH 15748 Eurycea bislineata 4.864 6.013 0.966 4.055 7.556 1.425 0.962 28.567 32.072 1.3 Non-cave_Aquatic
AMNH 32864 Eurycea bislineata 4.647 6.225 0.926 4.881 7.393 1.226 0.801 37.494 33.456 1.464 Non-cave_Aquatic
AMNH 51541 Eurycea bislineata 4.809 5.091 1.369 5.515 7.849 1.667 0.943 34.418 36.436 1.189 Non-cave_Aquatic
AMNH 51697 Eurycea bislineata 4.669 4.384 1.371 4.013 6.545 1.395 1.028 40.424 25.77 1.169 Non-cave_Aquatic
AMNH 60780 Eurycea bislineata 5.091 6.272 1.333 6.8 8.236 1.674 0.773 33.587 38.623 1.266 Non-cave_Aquatic
AMNH 60781 Eurycea bislineata 5.023 5.48 1.32 6.444 8.652 1.618 1.136 32.311 38.273 1.492 Non-cave_Aquatic
AMNH 116809 Eurycea bislineata 5.729 6.862 1.643 6.763 7.903 2.129 0.898 54.252 46.117 2.483 Non-cave_Aquatic
AMNH 116810 Eurycea bislineata 5.17 7.504 1.164 4.812 7.909 1.641 1.073 49.009 39.356 2.043 Non-cave_Aquatic
AMNH 116811 Eurycea bislineata 5.758 6.83 1.672 5.973 9.356 1.846 1.36 61.378 44.261 1.882 Non-cave_Aquatic
AMNH 116812 Eurycea bislineata 4.856 7.55 1.14 5.405 8.238 1.75 0.841 44.381 36.811 1.797 Non-cave_Aquatic 40
AMNH 116813 Eurycea bislineata 4.5 5.956 1.02 4.195 7.736 1.353 1.106 41.202 30.659 1.813 Non-cave_Aquatic
AMNH 164483 Eurycea bislineata 5.974 7.377 1.338 6.129 8.612 1.873 1.319 51.906 41.782 1.725 Non-cave_Aquatic
AMNH 164484 Eurycea bislineata 5.896 7.734 1.39 5.859 10.23 1.804 1.442 41.826 42.704 1.34 Non-cave_Aquatic
AMNH 44200 Eurycea cirrigera 6.152 7.166 1.918 7.337 11.241 2.389 1.498 60.864 41.973 4.241 Non-cave_Aquatic
AMNH 50866 Eurycea cirrigera 4.98 5.362 1.363 5.139 6.78 1.749 1.042 53.248 34.472 3.122 Non-cave_Aquatic
AMNH 50867 Eurycea cirrigera 5.502 7.267 1.21 6.571 8.297 1.667 0.737 38.319 38.065 2.032 Non-cave_Aquatic
AMNH 50869 Eurycea cirrigera 5.766 7.072 1.42 5.899 9.201 2.149 1.059 47.309 40.323 2.079 Non-cave_Aquatic
AMNH 60784 Eurycea cirrigera 5.09 5.826 1.048 4.899 6.684 1.771 0.526 42.854 32.216 1.453 Non-cave_Aquatic
AMNH 62827 Eurycea cirrigera 5.907 7.028 1.509 6.377 9.068 2.075 1.341 45.157 37.28 1.812 Non-cave_Aquatic
AMNH 100411 Eurycea cirrigera 6.146 7.207 1.412 5.751 8.256 2.148 1.102 32.231 36.042 2.786 Non-cave_Aquatic
AMNH 100412 Eurycea cirrigera 7.018 8.088 1.51 6.48 9.04 1.943 1.503 51.787 40.952 1.641 Non-cave_Aquatic
AMNH 100415 Eurycea cirrigera 5.614 7.959 1.034 5.067 9.726 1.306 0.981 42.946 36.996 2.18 Non-cave_Aquatic
AMNH 135499 Eurycea cirrigera 3.458 4.606 0.631 3.389 5.775 0.841 0.619 39.811 27.57 1.027 Non-cave_Aquatic
AMNH 135500 Eurycea cirrigera 3.477 4.579 0.756 3.243 5.953 0.994 0.631 26.641 25.499 1.367 Non-cave_Aquatic
AMNH 135501 Eurycea cirrigera 3.304 3.871 0.657 3.013 6.302 1.143 0.624 25.623 24.329 1.172 Non-cave_Aquatic
AMNH 143130 Eurycea cirrigera 5.224 5.798 1.147 4.924 7.457 1.728 1.158 50.784 33.427 1.803 Non-cave_Aquatic
AMNH 143131 Eurycea cirrigera 5.108 6.705 1.116 4.81 9.243 1.656 1.43 45.408 30.927 1.705 Non-cave_Aquatic
AMNH 143133 Eurycea cirrigera 4.02 4.122 0.656 3.649 4.57 1.029 0.582 42.314 25.825 1.278 Non-cave_Aquatic
AMNH 143134 Eurycea cirrigera 3.648 4.666 0.978 3.833 5.866 1.02 0.876 35.24 26.397 1.626 Non-cave_Aquatic
AMNH 143135 Eurycea cirrigera 5.302 6.008 1.277 4.293 7.114 1.377 1.028 24.618 30.77 1.692 Non-cave_Aquatic
AMNH 143136 Eurycea cirrigera 6.006 7.098 1.504 5.547 9.602 2.025 0.927 48.744 36.11 1.81 Non-cave_Aquatic
AMNH 182150 Eurycea cirrigera 5.292 5.457 1.306 6.614 8.003 2.029 1.179 51.56 38.85 1.766 Non-cave_Aquatic
AMNH 182151 Eurycea cirrigera 5.329 7.499 1.197 5.656 8.309 1.684 1.255 44.949 39.814 2.495 Non-cave_Aquatic
AMNH 182154 Eurycea cirrigera 4.941 6.313 1.15 4.707 8.222 1.531 0.943 38.861 33.243 2.11 Non-cave_Aquatic
AMNH 182155 Eurycea cirrigera 5.364 5.753 1.075 4.623 6.151 1.496 0.953 37.393 32.596 1.69 Non-cave_Aquatic
AMNH 182156 Eurycea cirrigera 5.967 8.115 1.145 6.175 8.897 1.822 1.358 50.585 36.248 2.49 Non-cave_Aquatic
AMNH 3090 Eurycea guttolineata 8.439 8.283 1.747 10.291 12.06 2.838 1.386 81.382 58.541 2.308 Non-cave_Terrestrial
AMNH 3997 Eurycea guttolineata 8.136 12.874 2.239 8.484 13.201 2.905 2.312 85.061 50.057 3.011 Non-cave_Terrestrial 41
AMNH 3998 Eurycea guttolineata 6.696 11.197 2.221 9.107 12.703 3.143 1.656 88.048 54.398 3.028 Non-cave_Terrestrial
AMNH 3999 Eurycea guttolineata 8.637 12.55 2.255 9.023 15.072 3.129 2.645 86.514 59.065 2.421 Non-cave_Terrestrial
AMNH 21195 Eurycea guttolineata 7.297 11.377 1.778 7.161 13.229 2.597 1.683 81.635 52.149 1.646 Non-cave_Terrestrial
AMNH 127073 Eurycea guttolineata 8.329 11.91 2.613 9.513 13.81 3.583 1.607 93.552 52.956 2.686 Non-cave_Terrestrial
AMNH 127075 Eurycea guttolineata 9.094 11.847 2.147 10.526 14.692 4.21 1.285 92.233 59.267 2.652 Non-cave_Terrestrial
AMNH 127076 Eurycea guttolineata 9.283 11.814 2.316 9.965 14.93 2.991 2.102 67.39 58.518 2.602 Non-cave_Terrestrial
AMNH 127087 Eurycea guttolineata 7.21 10.188 1.648 6.442 11.553 2.363 1.252 52.617 45.636 2.057 Non-cave_Terrestrial
AMNH 127088 Eurycea guttolineata 8.519 11.398 2.199 9.631 7.951 2.83 1.465 103.749 54.979 2.823 Non-cave_Terrestrial
AMNH 127089 Eurycea guttolineata 7.837 11.338 2.012 9.484 12.812 2.943 1.556 102.123 55.773 2.557 Non-cave_Terrestrial
AMNH 127092 Eurycea guttolineata 8.259 11.553 1.957 9.261 12.478 3.06 2.084 99.577 54.318 2.368 Non-cave_Terrestrial
AMNH 127093 Eurycea guttolineata 8.874 11.895 2.292 9.4 11.902 3.154 1.356 87.293 53.873 2.777 Non-cave_Terrestrial
AMNH 127102 Eurycea guttolineata 10.264 11.264 2.452 9.774 15.086 3.625 2.021 71.686 58.569 2.912 Non-cave_Terrestrial
AMNH 182160 Eurycea guttolineata 6.978 9.025 1.738 5.926 12.176 2.002 1.55 55.687 41.405 2.579 Non-cave_Terrestrial
AMNH 182161 Eurycea guttolineata 6.363 7.386 1.39 5.892 9.847 1.709 1.453 55.099 38.238 2.828 Non-cave_Terrestrial
AMNH 187780 Eurycea guttolineata 7.818 9.88 2.006 9.165 15.731 2.845 2.616 80.138 54.632 2.821 Non-cave_Terrestrial
AMNH 187781 Eurycea guttolineata 6.873 8.28 1.754 7.448 10.079 2.207 1.466 34.516 39.863 1.975 Non-cave_Terrestrial
AMNH 187782 Eurycea guttolineata 7.32 12.39 1.66 7.126 13.978 1.979 1.741 56.304 46.076 2.528 Non-cave_Terrestrial
AMNH 187783 Eurycea guttolineata 8.767 13.635 2.275 9.529 15.032 3.151 2.192 94.428 55.381 2.653 Non-cave_Terrestrial
AMNH 187784 Eurycea guttolineata 8 12.136 1.836 8.485 13.779 3.113 1.626 51.937 53.589 2.192 Non-cave_Terrestrial
AMNH 171578 Eurycea junaluska 4.141 5.944 0.778 3.748 7.576 1.357 0.646 43.76 30.914 1.443 Non-cave_Terrestrial
AMNH 172184 Eurycea junaluska 5.801 9.147 0.997 6.429 9.713 1.605 1.168 34.19 40.11 1.798 Non-cave_Terrestrial
AMNH 38057 Eurycea l. longicauda 6.859 11.758 1.84 5.537 14.402 1.981 1.203 87.513 54.983 2.149 Non-cave_Terrestrial
AMNH 43662 Eurycea l. longicauda 6.109 7.662 1.428 5.414 11.561 1.885 1.747 73.492 41.951 2.78 Non-cave_Terrestrial
AMNH 51698 Eurycea l. longicauda 7.633 13.903 2.212 10.087 15.2 3.356 1.607 93.134 51.571 2.542 Non-cave_Terrestrial
AMNH 58232 Eurycea l. longicauda 9.138 13.048 2.145 8.461 15.117 2.581 1.796 87.131 54.254 2.342 Non-cave_Terrestrial
AMNH 59810 Eurycea l. longicauda 7.391 10.025 1.433 5.951 14.01 2.387 0.794 61.405 51.295 2.188 Non-cave_Terrestrial
AMNH 79728 Eurycea l. longicauda 6.285 10.997 1.278 5.652 11.723 1.991 0.741 24.404 37.336 2.396 Non-cave_Terrestrial
AMNH 89783 Eurycea l. longicauda 8.072 13.612 1.769 8.154 16.433 2.409 1.919 101.849 57.656 2.919 Non-cave_Terrestrial 42
AMNH 99284 Eurycea l. longicauda 7.746 11.152 2.268 8.14 15.684 2.784 1.715 97.844 57.65 2.38 Non-cave_Terrestrial
AMNH 99286 Eurycea l. longicauda 8.415 14.088 1.76 6.858 18.345 2.386 1.458 59.503 62.019 2.513 Non-cave_Terrestrial
AMNH 112043 Eurycea l. longicauda 5.893 9.206 1.553 4.935 12.257 1.741 1.015 61.414 41.73 2.033 Non-cave_Terrestrial
AMNH 112044 Eurycea l. longicauda 5.75 9.066 1.25 5.344 9.953 2.057 1.041 61.321 40.555 1.82 Non-cave_Terrestrial
AMNH 112045 Eurycea l. longicauda 5.809 10.127 1.173 5.005 11.166 1.921 1.38 69.868 49.388 2.41 Non-cave_Terrestrial
AMNH 114012 Eurycea l. longicauda 6.994 9.902 1.986 8.222 12.442 2.682 1.496 88.872 54.015 1.983 Non-cave_Terrestrial
AMNH 114385 Eurycea l. longicauda 7.26 15.338 1.992 9.043 13.504 2.411 1.207 77.347 53.23 2.135 Non-cave_Terrestrial
AMNH 114386 Eurycea l. longicauda 5.82 10.936 1.397 5.748 12.109 2.419 1.041 61.822 39.017 1.959 Non-cave_Terrestrial
AMNH 115838 Eurycea l. longicauda 7.181 12.544 1.692 6.415 13.788 1.916 1.817 80.307 49.503 2.214 Non-cave_Terrestrial
AMNH 135240 Eurycea l. longicauda 7.924 14.599 1.816 8.067 15.588 2.837 1.598 87.949 58.252 3.157 Non-cave_Terrestrial
AMNH 135241 Eurycea l. longicauda 7.423 12.282 1.923 8.129 14.971 2.493 1.614 79.745 51.717 2.605 Non-cave_Terrestrial
AMNH 135782 Eurycea l. longicauda 6.495 12.608 1.861 6.806 15.264 2.426 2.079 93.709 48.254 2.535 Non-cave_Terrestrial
AMNH 136646 Eurycea l. longicauda 7.23 11.555 1.708 6.453 13.847 2.37 2.007 60.713 50.936 2.155 Non-cave_Terrestrial
AMNH 148944 Eurycea l. longicauda 6.659 9.453 1.481 5.831 12.119 1.719 1.447 55.706 42.587 2.453 Non-cave_Terrestrial
AMNH 148950 Eurycea l. longicauda 6.399 10.474 1.485 6.365 12.847 1.838 1.089 64.379 45.55 2.128 Non-cave_Terrestrial
AMNH 151305 Eurycea l. longicauda 7.889 12.968 1.826 7.231 13.499 2.235 1.737 74.055 54.626 2.333 Non-cave_Terrestrial
AMNH 40345 Eurycea l. melanopleura 7.62 10.747 1.718 6.213 13.113 2.111 1.937 74.188 49.066 2.499 Non-cave_Terrestrial
AMNH 40346 Eurycea l. melanopleura 8.626 11.891 1.638 6.858 11.772 2.796 1.577 80.155 44.461 2.467 Non-cave_Terrestrial
AMNH 40347 Eurycea l. melanopleura 7.692 9.978 1.522 6.396 12.388 2.021 1.102 64.487 47.485 2.343 Non-cave_Terrestrial
AMNH 40349 Eurycea l. melanopleura 6.261 10.283 1.236 6.171 10.886 1.757 1.708 36.223 40.633 2.125 Non-cave_Terrestrial
AMNH 40351 Eurycea l. melanopleura 7.402 10.454 1.919 7.08 12.816 2.195 1.388 82.262 49.768 2.672 Non-cave_Terrestrial
AMNH 40354 Eurycea l. melanopleura 6.273 8.259 1.519 5.715 11.478 1.664 1.1 62.701 55.342 2.259 Non-cave_Terrestrial
AMNH 40356 Eurycea l. melanopleura 8.712 9.951 1.524 7.28 12.056 2.576 1.685 83.172 45.669 2.478 Non-cave_Terrestrial
AMNH 52071 Eurycea l. melanopleura 9.054 8.712 2.087 8.768 12.798 2.697 1.959 83.032 54.316 2.458 Non-cave_Terrestrial
AMNH 52072 Eurycea l. melanopleura 8.617 12.106 2.318 9.463 13.303 3.309 1.55 53.551 56.266 2.787 Non-cave_Terrestrial
AMNH 52073 Eurycea l. melanopleura 8.293 11.984 2.082 10.27 15.429 2.973 1.802 86.971 57.069 2.657 Non-cave_Terrestrial
AMNH 52075 Eurycea l. melanopleura 7.898 9.522 1.524 7.377 12.393 2.441 1.175 60.643 43.999 2.048 Non-cave_Terrestrial
AMNH 52077 Eurycea l. melanopleura 9.649 12.335 2.349 10.192 13.62 3.174 1.954 74.178 58.875 2.789 Non-cave_Terrestrial 43
AMNH 52078 Eurycea l. melanopleura 8.505 12.655 1.903 9.481 14.589 2.927 1.554 67.296 60.688 2.6 Non-cave_Terrestrial
AMNH 59797 Eurycea l. melanopleura 6.142 9.774 1.086 5.621 11.682 1.595 1.523 54.255 41.083 2.363 Non-cave_Terrestrial
AMNH 59798 Eurycea l. melanopleura 5.64 7.349 1.182 4.93 10.199 1.595 0.8 40.804 36.57 1.671 Non-cave_Terrestrial
AMNH 59799 Eurycea l. melanopleura 6.239 8.882 1.307 4.872 11.034 1.704 1.301 32.847 40.588 2.259 Non-cave_Terrestrial
AMNH 59800 Eurycea l. melanopleura 5.588 8.096 1.03 4.83 9.752 1.321 0.895 41.353 35.52 2.297 Non-cave_Terrestrial
AMNH 149001 Eurycea l. melanopleura 5.213 8.581 1.055 4.727 9.625 1.698 1.069 44.608 35.411 2.033 Non-cave_Terrestrial
AMNH 149201 Eurycea l. melanopleura 7.794 11.172 1.656 7.241 12.367 2.551 1.262 61.83 45.241 2.271 Non-cave_Terrestrial
AMNH 149212 Eurycea l. melanopleura 8.786 11.507 1.956 10.484 13.034 2.795 1.684 87.046 53.768 2.568 Non-cave_Terrestrial
AMNH 149230 Eurycea l. melanopleura 7.97 12.033 2.284 9.056 13.588 3.044 1.566 77.246 56.029 2.873 Non-cave_Terrestrial
AMNH 149249 Eurycea l. melanopleura 7.532 11.992 2.02 7.827 12.347 2.511 1.197 73.182 48.489 2.452 Non-cave_Terrestrial
NMNH 123594 Eurycea latitans 7.991 6.943 1.546 7.115 8.871 1.895 1.477 45.758 46.945 1.15 Cave_Aquatic
NMNH 545379 Eurycea latitans 8.499 8.677 1.401 7.199 8.519 1.227 1.779 41.844 41.392 1.074 Cave_Aquatic
AMNH 11867 Eurycea lucifuga 8.835 13.762 2.078 7.478 17.943 2.627 1.742 72.338 53.867 2.407 Cave_Terrestrial
AMNH 11868 Eurycea lucifuga 9.114 14.758 2.387 8.402 20.227 2.47 2.772 87.605 63.245 1.943 Cave_Terrestrial
AMNH 16826 Eurycea lucifuga 10.045 17.462 2.189 8.95 19.346 2.529 2.191 78.219 64.906 3.057 Cave_Terrestrial
AMNH 22915 Eurycea lucifuga 6.289 10.881 1.473 4.878 9.346 1.288 1.562 60.8 37.313 2.654 Cave_Terrestrial
AMNH 22916 Eurycea lucifuga 8.262 16.264 2.227 9.052 16.463 2.993 1.963 80.656 55.416 3.519 Cave_Terrestrial
AMNH 22917 Eurycea lucifuga 8.599 13.923 1.873 7.828 13.071 2.066 2.324 70.119 46.463 3.164 Cave_Terrestrial
AMNH 22918 Eurycea lucifuga 5.751 8.802 1.125 5.171 10.6 2.305 1.44 22.33 39.554 1.945 Cave_Terrestrial
AMNH 22920 Eurycea lucifuga 8.752 13.1 2.084 8.638 14.213 2.783 2.228 82.81 55.115 3.156 Cave_Terrestrial
AMNH 22921 Eurycea lucifuga 6.841 12.186 1.75 5.777 11.454 2.104 1.884 45.598 62.327 2.337 Cave_Terrestrial
AMNH 22922 Eurycea lucifuga 7.085 10.437 1.161 5.575 11.326 2.071 1.731 56.376 42.017 2.34 Cave_Terrestrial
AMNH 32175 Eurycea lucifuga 6.137 9.108 1.222 4.474 10.765 1.796 1.199 44.7 35.17 2.103 Cave_Terrestrial
AMNH 32176 Eurycea lucifuga 5.73 8.213 1.201 4.819 10.41 1.682 1.023 43.592 35.055 2.049 Cave_Terrestrial
AMNH 33407 Eurycea lucifuga 9.573 15.588 2.152 7.873 15.226 2.797 1.741 83.79 60.548 2.759 Cave_Terrestrial
AMNH 33408 Eurycea lucifuga 8.628 14.404 2.035 6.82 14.613 2.577 1.36 83.463 57.146 2.769 Cave_Terrestrial
AMNH 33409 Eurycea lucifuga 8.642 14.66 2.128 7.829 16.127 2.889 2.182 83.135 56.52 2.973 Cave_Terrestrial
AMNH 36112 Eurycea lucifuga 5.479 6.601 1.201 4.549 8.843 1.704 0.814 19.463 35.544 2.299 Cave_Terrestrial 44
AMNH 36113 Eurycea lucifuga 5.666 7.943 1.208 4.785 9.452 1.899 1.255 42.509 34.784 2.058 Cave_Terrestrial
AMNH 36115 Eurycea lucifuga 5.854 9.905 1.269 4.624 10.76 1.331 1.012 39.087 35.959 1.893 Cave_Terrestrial
AMNH 36116 Eurycea lucifuga 5.895 8.599 1.193 5.216 10.132 1.673 1.039 17.884 36.321 2.342 Cave_Terrestrial
AMNH 36120 Eurycea lucifuga 6.144 10.216 1.487 4.942 9.36 1.926 1.129 36.264 36.771 2.069 Cave_Terrestrial
AMNH 38500 Eurycea lucifuga 10.297 17.352 2.507 8.234 19.074 2.435 2.231 63.567 62.796 2.562 Cave_Terrestrial
AMNH 40342 Eurycea lucifuga 9.337 16.878 2.469 7.99 18.258 2.556 1.618 64.874 60.08 3.005 Cave_Terrestrial
AMNH 41467 Eurycea lucifuga 6.947 12.175 1.817 6.946 12.443 2.964 1.951 70.418 47.096 1.851 Cave_Terrestrial
AMNH 41786 Eurycea lucifuga 8.038 13.998 1.702 6.476 16.564 2.31 2.816 37.623 49.144 2.382 Cave_Terrestrial
AMNH 50036 Eurycea lucifuga 6.985 12.247 1.569 6.095 13.82 2.147 1.319 62.894 44.728 1.895 Cave_Terrestrial
AMNH 52068 Eurycea lucifuga 10.278 16.933 3.042 10.61 15.891 3.282 1.658 87.868 64.13 3.395 Cave_Terrestrial
AMNH 52069 Eurycea lucifuga 8.639 14.282 1.736 7.631 14.167 2.619 1.814 72.62 51.597 3.566 Cave_Terrestrial
AMNH 52104 Eurycea lucifuga 10.491 16.401 2.258 10.586 17.421 3.174 1.99 86.751 61.561 3.053 Cave_Terrestrial
AMNH 52105 Eurycea lucifuga 9.854 15.691 2.75 10.368 15.457 3.523 1.805 80.444 57.544 3.246 Cave_Terrestrial
AMNH 52106 Eurycea lucifuga 10.251 15.926 2.153 7.403 18.124 3.525 2.005 68.362 61.454 3.404 Cave_Terrestrial
AMNH 52107 Eurycea lucifuga 10.173 15.134 2.688 7.26 19.516 2.968 2.141 90.776 65.173 3.727 Cave_Terrestrial
AMNH 52108 Eurycea lucifuga 9.806 15.095 1.865 8.91 15.143 3.194 1.444 89.484 63.845 3.486 Cave_Terrestrial
AMNH 52109 Eurycea lucifuga 9.674 15.36 2.13 9.039 18.062 2.765 1.543 86.221 59.878 3.459 Cave_Terrestrial
AMNH 52110 Eurycea lucifuga 10.704 13.747 2.104 9.621 15.946 3.178 1.88 98.029 63.134 3.653 Cave_Terrestrial
AMNH 52111 Eurycea lucifuga 10.397 12.453 1.953 7.084 14.422 3.004 1.71 63.746 54.043 3.113 Cave_Terrestrial
AMNH 52112 Eurycea lucifuga 8.406 12.906 1.959 7.27 13.592 2.628 1.836 76.394 51.894 3.257 Cave_Terrestrial
AMNH 52113 Eurycea lucifuga 9.035 14.87 1.901 7.787 14.286 2.353 2.083 75.554 57.213 3.886 Cave_Terrestrial
AMNH 52114 Eurycea lucifuga 7.847 12.663 1.513 6.243 15.845 2.734 2.395 69.316 51.308 3.138 Cave_Terrestrial
AMNH 52115 Eurycea lucifuga 8.653 14.162 1.919 6.493 13.455 2.762 1.574 87.37 54.992 3.824 Cave_Terrestrial
AMNH 52117 Eurycea lucifuga 9.951 17.052 2.243 8.677 17.232 2.711 1.826 65.411 57.671 3.354 Cave_Terrestrial
AMNH 52118 Eurycea lucifuga 9.046 12.949 1.926 8.073 13.624 1.786 2.118 75.353 55.074 2.315 Cave_Terrestrial
AMNH 52119 Eurycea lucifuga 9.602 16.117 2.283 8.855 16.591 2.938 1.678 89.114 62.928 3.454 Cave_Terrestrial
AMNH 52121 Eurycea lucifuga 9.743 16.259 2.429 8.228 19.049 3.449 1.987 65.683 60.547 2.875 Cave_Terrestrial
AMNH 52443 Eurycea lucifuga 10.743 17.62 2.661 8.708 18.105 3.555 1.957 73.129 72.792 3.286 Cave_Terrestrial 45
AMNH 52444 Eurycea lucifuga 10.357 13.532 2.728 7.985 14.888 2.745 1.838 75.317 60.682 3.608 Cave_Terrestrial
AMNH 52445 Eurycea lucifuga 10.248 13.398 2.194 10.52 16.237 3.181 1.348 49.48 58.432 4.019 Cave_Terrestrial
AMNH 52446 Eurycea lucifuga 11.948 16.155 3.284 11.944 15.76 4.098 2.489 95.947 68.362 3.668 Cave_Terrestrial
AMNH 52475 Eurycea lucifuga 9.555 12.817 2.156 8.004 15.557 3.29 1.684 90.28 57.608 3.556 Cave_Terrestrial
AMNH 52527 Eurycea lucifuga 9.038 13.866 1.661 5.681 15.204 1.903 1.991 61.356 44.06 2.201 Cave_Terrestrial
AMNH 54554 Eurycea lucifuga 11.179 14.243 2.55 7.484 17.492 3.168 1.493 58.82 68.111 3.705 Cave_Terrestrial
AMNH 54555 Eurycea lucifuga 9.748 15.288 2.157 7.053 15.356 2.411 1.561 81.683 62.531 3.829 Cave_Terrestrial
AMNH 56367 Eurycea lucifuga 7.081 11.155 1.206 5.408 13.487 1.936 1.668 52.959 44.481 2.434 Cave_Terrestrial
AMNH 56368 Eurycea lucifuga 9.97 18.553 2.136 8.033 19.41 2.738 1.854 65.803 64.546 2.924 Cave_Terrestrial
AMNH 58233 Eurycea lucifuga 9.843 16.616 1.92 9.065 21.288 3.119 2.327 111.287 61.433 2.798 Cave_Terrestrial
AMNH 58234 Eurycea lucifuga 9.627 14.66 2.246 7.74 17.59 2.924 2.926 112.952 60.326 2.778 Cave_Terrestrial
AMNH 58235 Eurycea lucifuga 9.706 16.969 1.763 7.695 17.375 2.949 2.652 97.394 57.321 2.982 Cave_Terrestrial
AMNH 58236 Eurycea lucifuga 7.199 14.036 2.215 7.212 15.273 2.873 2.362 83.224 49.416 2.616 Cave_Terrestrial
AMNH 59020 Eurycea lucifuga 9.949 16.082 2.142 7.899 14.404 2.294 2.254 100.55 62.842 3.609 Cave_Terrestrial
AMNH 59021 Eurycea lucifuga 9.577 11.977 2.144 8.923 13.735 2.777 2.285 95.046 58.132 3.096 Cave_Terrestrial
AMNH 59022 Eurycea lucifuga 8.323 12.185 2.028 6.822 16.135 2.6 2.096 83.843 54.214 3.174 Cave_Terrestrial
AMNH 59758 Eurycea lucifuga 11.37 16.852 2.448 8.768 16.981 3.076 1.445 87.448 66.974 3.381 Cave_Terrestrial
AMNH 59759 Eurycea lucifuga 9.307 15.717 1.663 7.019 18.186 2.619 2.306 62.508 52.743 3.089 Cave_Terrestrial
AMNH 59760 Eurycea lucifuga 9.685 15.949 2.489 7.204 16.144 2.464 1.948 45.587 60.544 3.139 Cave_Terrestrial
AMNH 59761 Eurycea lucifuga 8.92 15.7 2.109 7.063 18.241 2.487 2.673 89.711 59.922 2.915 Cave_Terrestrial
AMNH 59762 Eurycea lucifuga 7.506 11.725 1.669 5.282 14.393 1.861 1.598 61.957 45.279 2.43 Cave_Terrestrial
AMNH 59764 Eurycea lucifuga 11.543 16.953 2.635 9.708 18.043 3.287 3.113 79.978 62.149 3.125 Cave_Terrestrial
AMNH 59765 Eurycea lucifuga 8.696 11.597 1.93 7.039 13.93 2.152 1.76 74.496 55.809 2.905 Cave_Terrestrial
AMNH 59767 Eurycea lucifuga 11.14 17.152 2.499 9.193 16.625 3.04 1.27 76.619 61.347 3.035 Cave_Terrestrial
AMNH 59768 Eurycea lucifuga 8.309 14.032 1.721 6.803 15.873 2.292 1.933 77.071 54.986 2.666 Cave_Terrestrial
AMNH 59769 Eurycea lucifuga 8.67 14.313 1.984 8.171 16.535 2.687 1.584 90.558 58.845 3.246 Cave_Terrestrial
AMNH 59770 Eurycea lucifuga 10.702 15.783 2.314 7.951 16.923 2.65 2.598 100.495 71.033 3.348 Cave_Terrestrial
AMNH 59771 Eurycea lucifuga 10.17 15.257 2.206 7.253 16.894 2.845 2.543 76.438 59.669 2.763 Cave_Terrestrial 46
AMNH 59772 Eurycea lucifuga 8.507 13.026 1.813 6.025 13.548 2.352 1.934 65.271 53.001 2.801 Cave_Terrestrial
AMNH 59780 Eurycea lucifuga 11.349 13.46 2.057 9.513 18.492 2.911 2.17 90.315 58.543 3.014 Cave_Terrestrial
AMNH 59781 Eurycea lucifuga 12.442 17.857 3.02 12.245 17.289 4.107 2.615 84.226 66.225 3.765 Cave_Terrestrial
AMNH 59782 Eurycea lucifuga 10.478 14.301 2.661 10.406 17.016 2.903 1.666 68.78 59.79 3.183 Cave_Terrestrial
AMNH 59803 Eurycea lucifuga 8.874 13.955 8.144 6.166 9.843 2.428 1.754 78.01 55.488 3.164 Cave_Terrestrial
AMNH 59804 Eurycea lucifuga 9.575 17.61 2.076 8.774 18.407 3.063 1.84 70.975 65.827 3.298 Cave_Terrestrial
AMNH 59805 Eurycea lucifuga 11.045 17.125 2.399 8.515 16.043 3.219 2.196 81.658 64.328 3.391 Cave_Terrestrial
AMNH 59806 Eurycea lucifuga 10.402 14.001 2.382 8.31 17.322 3.372 1.701 81.556 61.264 3.365 Cave_Terrestrial
AMNH 59807 Eurycea lucifuga 9.513 15.933 2.39 8.618 18.279 3.248 1.904 91.709 59.594 3.388 Cave_Terrestrial
AMNH 137469 Eurycea lucifuga 9.785 18.695 2.085 9.018 17.12 2.706 2.373 77.406 65.295 3.534 Cave_Terrestrial
AMNH 137470 Eurycea lucifuga 10.119 15.641 2.003 8.227 20.534 2.357 2.336 98.909 66.436 3.528 Cave_Terrestrial
AMNH 137471 Eurycea lucifuga 9.758 18.251 1.487 7.804 19.245 2.498 2.446 82.319 63.379 2.932 Cave_Terrestrial
AMNH 137472 Eurycea lucifuga 10.462 17.448 2.443 8.883 20.597 2.574 2.546 89.321 65.703 2.457 Cave_Terrestrial
AMNH 137473 Eurycea lucifuga 9.583 16.614 1.423 6.607 16.783 2.25 2.695 46.575 60.052 3.043 Cave_Terrestrial
AMNH 143530 Eurycea lucifuga 8.386 12.359 2.001 7.814 14.892 2.94 2.093 74.072 53.145 2.682 Cave_Terrestrial
AMNH 143532 Eurycea lucifuga 10.463 15.148 2.152 9.828 16.696 2.84 2.453 44.834 64.299 3.116 Cave_Terrestrial
AMNH 143533 Eurycea lucifuga 8.025 12.146 1.495 6.425 15.303 2.094 1.56 82.282 54.298 2.001 Cave_Terrestrial
AMNH 143534 Eurycea lucifuga 11.342 15.318 2.426 8.268 18.668 4.103 2.118 71.429 62.393 3.532 Cave_Terrestrial
AMNH 143535 Eurycea lucifuga 7.701 12.23 1.552 6.355 15.195 2.798 1.643 53.655 48.355 2.409 Cave_Terrestrial
AMNH 143536 Eurycea lucifuga 8.836 13.615 1.588 6.5 15.323 2.572 1.925 75.323 58.023 2.38 Cave_Terrestrial
AMNH 143537 Eurycea lucifuga 9.026 14.207 1.953 7.592 15.52 2.657 2.054 58.387 50.401 3.259 Cave_Terrestrial
AMNH 143539 Eurycea lucifuga 9.979 16.261 2.514 7.97 16.788 3.136 2.021 58.636 62.492 3.543 Cave_Terrestrial
AMNH 143540 Eurycea lucifuga 10.03 15.801 2.238 7.711 17.335 2.858 2.418 83.866 58.229 3.204 Cave_Terrestrial
AMNH 143541 Eurycea lucifuga 10.909 17.206 2.814 8.283 18.966 3.75 1.32 36.301 58.08 3.677 Cave_Terrestrial
AMNH 143542 Eurycea lucifuga 10.483 13.325 2.025 7.933 14.178 3.251 2.056 74.078 57.939 3.202 Cave_Terrestrial
AMNH 143543 Eurycea lucifuga 9.08 14.55 2.138 8.493 17.079 2.902 1.924 47.94 56.601 2.912 Cave_Terrestrial
AMNH 143545 Eurycea lucifuga 10.735 12.88 2.309 8.474 17.499 3.09 2.045 65.707 63.167 2.539 Cave_Terrestrial
AMNH 143547 Eurycea lucifuga 9.182 11.639 1.805 8.366 13.759 2.414 2.338 82.181 59.8 2.745 Cave_Terrestrial 47
AMNH 143548 Eurycea lucifuga 10.002 16.141 2.095 8.379 17.307 3.759 1.434 93.631 59.452 1.926 Cave_Terrestrial
AMNH 143549 Eurycea lucifuga 9.882 14.817 2.166 7.482 14.943 3.223 2.063 44.798 62.504 2.75 Cave_Terrestrial
AMNH 143552 Eurycea lucifuga 8.214 12.744 1.876 8.514 13.741 2.63 2.515 81.115 57.078 2.737 Cave_Terrestrial
AMNH 143553 Eurycea lucifuga 6.574 14.214 1.95 7.743 13.677 1.891 1.902 81.083 57.566 4.087 Cave_Terrestrial
AMNH 143554 Eurycea lucifuga 8.041 12.743 1.787 6.697 17.589 2.362 2.464 73.88 50.818 2.927 Cave_Terrestrial
AMNH 143555 Eurycea lucifuga 7.85 12.4 1.957 6.718 14.624 2.399 1.626 59.124 50.298 2.468 Cave_Terrestrial
AMNH 143556 Eurycea lucifuga 6.589 9.574 0.978 5.36 13.938 2.043 1.609 71.062 49.159 Cave_Terrestrial
AMNH 143557 Eurycea lucifuga 8.955 13.147 2.549 7.755 18.593 2.973 1.561 50.697 49.244 3.013 Cave_Terrestrial
AMNH 143558 Eurycea lucifuga 10.425 12.598 1.904 8.034 11.013 3.448 1.896 89.405 61.082 3.802 Cave_Terrestrial
AMNH 143559 Eurycea lucifuga 7.832 14.936 1.333 6.034 17.038 1.885 2.117 88.793 56.807 3.632 Cave_Terrestrial
AMNH 143560 Eurycea lucifuga 9.462 16.111 1.825 7.624 16.25 3.368 2.241 86.722 58.721 3.515 Cave_Terrestrial
AMNH 143561 Eurycea lucifuga 10.039 15.521 2.329 8.375 15.871 3.504 2.373 78.51 57.3 3.176 Cave_Terrestrial
AMNH 143562 Eurycea lucifuga 8.167 11.716 1.83 6.747 14.169 2.735 1.191 73.148 51.369 2.514 Cave_Terrestrial
AMNH 143563 Eurycea lucifuga 10.271 13.971 1.986 8.794 15.583 3.216 2.04 88.6 59.777 2.585 Cave_Terrestrial
AMNH 143564 Eurycea lucifuga 7.97 15.332 1.735 5.886 15.421 3.107 2.039 72.532 51.914 2.585 Cave_Terrestrial
AMNH 143565 Eurycea lucifuga 9.258 12.581 1.694 7.139 13.251 2.812 1.457 73.747 56.157 2.879 Cave_Terrestrial
AMNH 143566 Eurycea lucifuga 7.393 13.879 2.062 6.73 15.838 3.111 2.187 70.714 48.198 3.176 Cave_Terrestrial
AMNH 143567 Eurycea lucifuga 8.658 13.403 1.834 4.439 9.598 2.866 1.57 65.834 48.489 2.586 Cave_Terrestrial
AMNH 143568 Eurycea lucifuga 8.85 12.962 2.313 7.242 15.296 2.711 1.591 74.059 54.036 3.303 Cave_Terrestrial
AMNH 143569 Eurycea lucifuga 7.312 10.6 1.839 6.537 15.152 2.725 1.675 67.174 45.214 2.475 Cave_Terrestrial
AMNH 143570 Eurycea lucifuga 9.894 16.72 2.323 8.365 18.454 2.87 1.508 69.779 56.122 3.265 Cave_Terrestrial
AMNH 143573 Eurycea lucifuga 9.644 12.236 2.447 7.002 15.183 2.576 1.619 89.082 53.277 3.242 Cave_Terrestrial
AMNH 143574 Eurycea lucifuga 9.434 13.015 1.808 7.142 16.121 2.452 1.622 79.805 53.557 3.017 Cave_Terrestrial
AMNH 143575 Eurycea lucifuga 8.096 11.406 1.877 5.959 15.389 2.308 1.598 58.546 48.468 2.74 Cave_Terrestrial
AMNH 143576 Eurycea lucifuga 8.779 13.452 1.841 6.018 14.972 2.553 1.597 72.889 50.533 2.494 Cave_Terrestrial
AMNH 143577 Eurycea lucifuga 9.033 13.316 2.07 7.64 17.159 2.796 1.298 60.537 53.441 2.796 Cave_Terrestrial
AMNH 143578 Eurycea lucifuga 10.283 15.83 2.37 8.68 16.283 3.241 1.82 77.822 61.63 2.45 Cave_Terrestrial
AMNH 143579 Eurycea lucifuga 10.546 16.943 2.637 9.462 18.707 3.033 1.82 99.526 59.269 2.626 Cave_Terrestrial 48
AMNH 143580 Eurycea lucifuga 9.653 14.189 2.398 7.195 16.168 3.049 1.974 87.148 61.679 3.115 Cave_Terrestrial
AMNH 143582 Eurycea lucifuga 7.807 12.654 1.768 6.244 13.902 2.221 1.583 51.365 46.775 2.315 Cave_Terrestrial
AMNH 143583 Eurycea lucifuga 9.582 16.395 2.178 7.391 15.697 2.884 2.249 21.468 55.945 3.617 Cave_Terrestrial
AMNH 143584 Eurycea lucifuga 8.378 15.014 1.727 7.195 14.738 2.334 1.837 72.467 50.98 3.622 Cave_Terrestrial
AMNH 143585 Eurycea lucifuga 7.745 12.745 1.65 5.28 14.466 1.788 1.626 32.056 49.879 2.783 Cave_Terrestrial
AMNH 143588 Eurycea lucifuga 6.928 11.422 1.187 5.359 11.242 1.835 1.355 26.582 43.049 2.476 Cave_Terrestrial
AMNH 143589 Eurycea lucifuga 10.32 16.165 2.447 7.715 18.073 3.563 1.948 47.928 60.009 3.936 Cave_Terrestrial
AMNH 155760 Eurycea lucifuga 10.837 14.761 2.78 8.583 18.214 3.438 1.775 100.342 65.176 2.822 Cave_Terrestrial
AMNH 155761 Eurycea lucifuga 9.833 16.186 2.138 9.522 17.33 2.72 1.864 88.877 60.004 2.982 Cave_Terrestrial
AMNH 182172 Eurycea lucifuga 8.97 15.588 2.196 8.599 16.512 2.84 2.283 75.127 55.547 3.08 Cave_Terrestrial
AMNH 182173 Eurycea lucifuga 7.836 9.328 1.842 8.731 15.839 2.919 2.743 80.571 55.312 2.543 Cave_Terrestrial
AMNH 187787 Eurycea lucifuga 8.655 12.518 1.617 7.666 13.602 2.062 2.5222 71.617 54.139 1.489 Cave_Terrestrial
AMNH 187788 Eurycea lucifuga 10.953 14.363 2.188 9.298 15.391 3.256 2.166 71.541 66.2 2.668 Cave_Terrestrial
AMNH 187789 Eurycea lucifuga 8.032 13.29 1.665 6.528 15.495 2.684 1.854 52.427 51.987 3.321 Cave_Terrestrial
AMNH 11865 Eurycea multiplicata 4.354 6.491 1.02 5.009 6.727 1.092 0.516 29.727 45.457 1.232 Non-cave_Aquatic
AMNH 32100 Eurycea multiplicata 4.78 6.747 1.143 4.495 7.688 0.936 0.679 22.446 33.738 1.456 Non-cave_Aquatic
AMNH 40358 Eurycea multiplicata 8.954 11.331 2.207 10.124 14.946 3.071 1.713 87.019 55.592 2.959 Non-cave_Aquatic
AMNH 40360 Eurycea multiplicata 5.039 5.629 0.902 3.741 7.338 1.49 0.567 37.834 39.348 1.161 Non-cave_Aquatic
AMNH 40361 Eurycea multiplicata 5.32 6.945 1.294 5.286 9.274 1.653 0.867 32.81 48.767 1.241 Non-cave_Aquatic
AMNH 40362 Eurycea multiplicata 5.476 6.584 0.949 4.613 8.236 1.358 0.701 23.828 42.503 1.455 Non-cave_Aquatic
AMNH 40363 Eurycea multiplicata 5.174 5.024 1.277 5.595 7.978 1.582 0.787 38.617 39.142 1.071 Non-cave_Aquatic
AMNH 52461 Eurycea multiplicata 4.014 4.943 0.778 4.127 6.104 1.146 0.919 38.646 34.283 0.836 Non-cave_Aquatic
AMNH 52462 Eurycea multiplicata 4.251 5.217 0.928 4.836 5.651 1.083 0.845 25.241 29.515 1.35 Non-cave_Aquatic
AMNH 59813 Eurycea multiplicata 5.724 7.197 1.209 6.445 8.046 1.595 1.533 43.672 42.993 1.539 Non-cave_Aquatic
AMNH 143651 Eurycea multiplicata 6.275 7.502 1.252 5.899 9.353 1.604 1.014 36.441 43.965 1.583 Non-cave_Aquatic
AMNH 151182 Eurycea multiplicata 6.087 4.948 1.119 5.605 6.609 1.538 0.724 31.773 35.056 1.399 Non-cave_Aquatic
AMNH 151191 Eurycea multiplicata 4.402 4.435 0.657 4.537 5.161 0.875 0.692 20.003 28.072 1.106 Non-cave_Aquatic
AMNH 151197 Eurycea multiplicata 4.612 3.678 0.78 3.555 4.343 0.714 0.512 18.615 24.05 0.859 Non-cave_Aquatic 49
AMNH 151199 Eurycea multiplicata 4.553 6.243 1.169 4.155 6.226 1.143 0.563 29.565 37.328 1.441 Non-cave_Aquatic
AMNH 151200 Eurycea multiplicata 5.549 7.118 1.153 5.155 7.411 1.408 0.709 43.643 35.579 1.514 Non-cave_Aquatic
AMNH 151201 Eurycea multiplicata 4.659 5.967 1.149 3.974 7.376 1.286 0.677 32.258 33.998 1.324 Non-cave_Aquatic
AMNH 151202 Eurycea multiplicata 4.561 5.988 0.73 3.641 6.972 1.056 0.856 19.928 25.097 1.28 Non-cave_Aquatic
AMNH 182174 Eurycea multiplicata 5.815 6.828 1.34 6.09 7.238 1.456 0.657 42.875 41.026 1.826 Non-cave_Aquatic
AMNH 182175 Eurycea multiplicata 5.159 5.288 1.118 5.799 5.862 1.594 0.868 40.177 39.775 1.577 Non-cave_Aquatic
AMNH 60789 Eurycea nana 2.986 3.264 0.48 2.347 5.164 0.821 0.596 19.642 27.167 1.001 Non-cave_Aquatic
AMNH 108768 Eurycea nana 3.882 5.843 0.601 4.271 5.877 0.902 0.664 19.53 29.33 1.143 Non-cave_Aquatic
AMNH 108769 Eurycea nana 3.368 3.89 0.594 3.537 3.359 0.629 0.52 20.229 24.692 0.991 Non-cave_Aquatic
AMNH 108771 Eurycea nana 3.638 4.533 0.602 2.824 4.949 0.613 0.424 20.188 22.547 0.707 Non-cave_Aquatic
AMNH 108772 Eurycea nana 3.417 5.002 0.516 2.73 4.618 0.626 0.34 20.988 27.376 1.199 Non-cave_Aquatic
AMNH 108773 Eurycea nana 3.723 5.191 0.381 2.287 6.105 0.558 0.774 24.302 28.077 1.419 Non-cave_Aquatic
AMNH 108774 Eurycea nana 3.613 4.718 0.586 3.562 4.78 0.72 0.809 20.604 25.641 0.999 Non-cave_Aquatic
AMNH 108776 Eurycea nana 3.617 4.129 0.503 2.773 4.515 0.803 0.566 17.902 22.201 0.951 Non-cave_Aquatic
AMNH 108778 Eurycea nana 3.122 3.718 0.376 2.476 4.01 0.493 0.438 18.686 25.354 1.14 Non-cave_Aquatic
AMNH 108779 Eurycea nana 3.235 3.815 0.467 2.206 4.426 0.59 0.571 15.101 19.991 0.908 Non-cave_Aquatic
AMNH 182176 Eurycea nana 3.32 3.12 0.489 3.161 3.91 0.64 0.35 23.853 31.601 1.003 Non-cave_Aquatic
AMNH 182177 Eurycea nana 2.971 2.73 0.336 1.935 4.549 0.397 0.401 14.66 24.445 0.813 Non-cave_Aquatic
AMNH 182178 Eurycea nana 2.976 2.653 0.35 2.36 3.242 0.357 0.306 16.872 23.01 0.814 Non-cave_Aquatic
AMNH 182179 Eurycea nana 3.015 3.206 0.327 2.273 3.566 0.495 0.347 18.317 23.455 0.727 Non-cave_Aquatic
NMNH 545578 Eurycea nana 3.942 4.365 0.671 4.111 4.805 0.663 1.18 24.67 25.87 0.901 Non-cave_Aquatic
NMNH 545579 Eurycea nana 3.85 4.328 0.557 3.774 4.751 0.593 0.658 23.052 29.037 0.739 Non-cave_Aquatic
NMNH 545581 Eurycea nana 3.459 3.624 0.428 2.733 3.445 0.613 0.883 16.847 21.704 0.902 Non-cave_Aquatic
NMNH 545584 Eurycea nana 3.849 4.198 0.726 3.413 4.944 0.763 0.859 24.114 24.876 0.966 Non-cave_Aquatic
NMNH 545585 Eurycea nana 3.717 4.445 0.561 2.654 4.42 0.616 1.063 16.24 25.344 1.033 Non-cave_Aquatic
NMNH 545587 Eurycea nana 3.526 3.725 0.566 2.78 3.862 0.693 0.754 17.544 21.285 0.64 Non-cave_Aquatic
NMNH 545588 Eurycea nana 3.455 3.387 0.408 3.009 3.648 0.458 0.55 17.417 19.928 0.834 Non-cave_Aquatic
NMNH 545589 Eurycea nana 3.343 3.643 0.408 3.31 4.756 0.479 0.822 19.483 19.707 0.701 Non-cave_Aquatic 50
AMNH 62054 Eurycea neotenes 4.413 6.355 0.791 3.876 7.014 0.89 0.805 27.319 31.503 0.901 Non-cave_Aquatic
AMNH 62057 Eurycea neotenes 4.267 5.273 0.855 3.45 5.815 0.9 0.478 23.643 32.456 1.126 Non-cave_Aquatic
AMNH 62058 Eurycea neotenes 3.797 4.597 0.87 3.011 5.412 0.895 0.698 16.88 23.318 1.232 Non-cave_Aquatic
AMNH 188124 Eurycea neotenes 4.441 5.759 0.959 5.434 6.721 0.994 0.618 31.609 35.816 1.138 Non-cave_Aquatic
AMNH 188126 Eurycea neotenes 4.474 4.929 0.729 3.432 5.57 0.808 0.642 22.39 27.117 0.849 Non-cave_Aquatic
AMNH 188127 Eurycea neotenes 5.318 6.163 0.934 4.463 6.899 1.032 0.675 26.234 36.746 1.173 Non-cave_Aquatic
AMNH 188128 Eurycea neotenes 4.612 5.763 0.656 3.941 5.792 1.034 0.536 26.781 31.19 1.026 Non-cave_Aquatic
AMNH 188129 Eurycea neotenes 4.174 5.522 0.627 4.139 6.166 0.696 0.581 29.217 32.044 1.028 Non-cave_Aquatic
AMNH 188130 Eurycea neotenes 4.792 5.857 1.021 4.846 4.551 1.163 0.809 24.549 31.537 1.226 Non-cave_Aquatic
AMNH 188133 Eurycea neotenes 4.751 5.524 0.699 4.385 6.061 1.081 0.508 28.723 34.204 0.727 Non-cave_Aquatic
AMNH 188134 Eurycea neotenes 3.666 3.985 0.45 3.959 5.51 0.956 0.628 18.375 26.718 0.65 Non-cave_Aquatic
AMNH 188137 Eurycea neotenes 4.316 5.227 0.861 4.166 5.205 0.65 0.775 24.528 33.44 0.795 Non-cave_Aquatic
AMNH 188138 Eurycea neotenes 4.296 4.019 0.598 3.799 3.672 0.637 0.549 20.881 26.094 0.846 Non-cave_Aquatic
AMNH 188139 Eurycea neotenes 5.276 5.353 0.834 4.311 6.129 0.933 0.462 25.186 32.159 1.065 Non-cave_Aquatic
AMNH 188145 Eurycea neotenes 5.43 5.677 0.674 4.332 5.998 0.774 0.504 25.324 35.402 0.785 Non-cave_Aquatic
AMNH 188150 Eurycea neotenes 4.429 3.347 0.664 4.815 4.675 0.9 0.47 24.182 29.331 0.827 Non-cave_Aquatic
AMNH 188151 Eurycea neotenes 4.714 4.733 0.752 4.325 5.967 0.773 0.49 23.555 29.318 0.496 Non-cave_Aquatic
AMNH 188172 Eurycea neotenes 3.738 3.956 0.569 3.424 4.509 0.842 0.345 20.232 25.652 0.713 Non-cave_Aquatic
AMNH 188198 Eurycea neotenes 4.991 5.099 0.961 6.135 7.017 0.866 0.954 25.301 36.295 1.09 Non-cave_Aquatic
AMNH 188199 Eurycea neotenes 5.853 6.264 0.993 5.131 6.411 1.2 0.996 26.958 35.955 1.067 Non-cave_Aquatic
AMNH 188200 Eurycea pterophila 5.911 5.176 1.194 5.35 4.985 1.358 0.782 28.249 34.534 0.967 Cave_Aquatic
AMNH 44336 Eurycea quadridigitata 4.828 5.844 0.803 5.204 7.532 1.128 0.926 51.089 35.927 1.381 Non-cave_Terrestrial
AMNH 44337 Eurycea quadridigitata 3.627 4.051 0.963 4.226 6.384 1.167 0.5 29.046 29.681 0.775 Non-cave_Terrestrial
AMNH 53905 Eurycea quadridigitata 3.848 5.036 0.55 3.78 6.018 1.036 0.603 46.393 30.356 1.053 Non-cave_Terrestrial
AMNH 72593 Eurycea quadridigitata 3.817 5.975 0.745 4.101 7.549 1.105 0.826 36.709 28.646 1.465 Non-cave_Terrestrial
AMNH 89818 Eurycea quadridigitata 3.421 4.879 0.533 2.91 6.565 0.807 0.881 31.827 26.088 1.075 Non-cave_Terrestrial
AMNH 89821 Eurycea quadridigitata 4.334 5.826 0.968 5.19 6.495 1.246 0.772 48.685 33.625 1.557 Non-cave_Terrestrial
AMNH 89822 Eurycea quadridigitata 4.217 6.459 0.866 4.352 7.419 1.149 0.925 53.357 34.102 1.47 Non-cave_Terrestrial 51
AMNH 89828 Eurycea quadridigitata 4.04 5.009 0.882 4.944 6.397 1.151 0.806 41.638 29.982 1.161 Non-cave_Terrestrial
AMNH 89834 Eurycea quadridigitata 3.985 5.4 0.698 4.061 7.704 1.092 0.848 47.968 32.708 1.444 Non-cave_Terrestrial
AMNH 89835 Eurycea quadridigitata 3.24 4.664 0.735 3.839 5.792 1.086 0.833 36.601 28.894 1.172 Non-cave_Terrestrial
AMNH 93055 Eurycea quadridigitata 4.078 3.639 0.98 3.969 6.284 1.13 0.555 37.025 27.249 1.118 Non-cave_Terrestrial
AMNH 93058 Eurycea quadridigitata 3.697 3.735 0.859 3.466 5.011 0.847 0.33 31.966 26.554 1.556 Non-cave_Terrestrial
AMNH 125819 Eurycea quadridigitata 4.986 5.829 0.911 5.222 6.839 1.152 0.799 49.504 30.402 1.188 Non-cave_Terrestrial
AMNH 143803 Eurycea quadridigitata 4.119 5.658 0.944 3.864 6.729 1.314 0.778 40.532 28.052 1.388 Non-cave_Terrestrial
AMNH 172401 Eurycea quadridigitata 4.296 6.037 0.928 4.229 7.842 1.166 0.909 45.762 35.188 1.385 Non-cave_Terrestrial
AMNH 172404 Eurycea quadridigitata 4.418 6.987 0.85 4.625 7.9 1.13 1.199 39.295 37.502 1.58 Non-cave_Terrestrial
AMNH 182183 Eurycea quadridigitata 3.76 4.107 1.03 4.726 6.88 1.159 0.785 40.634 31.121 1.779 Non-cave_Terrestrial
AMNH 188213 Eurycea quadridigitata 4.443 6.428 0.945 4.113 6.708 1.218 0.926 41.036 28.112 1.213 Non-cave_Terrestrial
AMNH 188214 Eurycea quadridigitata 4.275 5.769 0.923 3.828 6.947 1.007 0.861 41.612 29.177 1.498 Non-cave_Terrestrial
AMNH 188215 Eurycea quadridigitata 4.029 5.912 0.771 4.516 7.007 0.987 1.152 40.07 27.14 1.307 Non-cave_Terrestrial
AMNH 188216 Eurycea quadridigitata 4.889 6.481 1.113 5.582 7.747 1.231 0.704 44.672 34.063 1.325 Non-cave_Terrestrial
AMNH 2276 Eurycea rathbuni 15.178 16.763 1.585 8.817 15.741 2.083 1.228 41.758 60.874 0.788 Cave_Aquatic
AMNH 2279 Eurycea rathbuni 12.573 14.845 1.436 8.167 17.606 2.811 0.998 33.12 66.526 0.503 Cave_Aquatic
AMNH 2281 Eurycea rathbuni 13.107 13.681 1.613 7.797 14.035 1.943 0.73 25.253 61.258 0.694 Cave_Aquatic
AMNH 2282 Eurycea rathbuni 10.741 13.084 1.253 6.995 12.778 1.444 0.852 30.563 47.601 0.337 Cave_Aquatic
AMNH 2285 Eurycea rathbuni 13.017 9.768 1.288 7.297 15.52 2.5 0.755 32.738 52.65 0.77 Cave_Aquatic
AMNH 2288 Eurycea rathbuni 9.153 10.799 1.359 6.135 8.997 1.112 1.684 23.054 39.645 0.409 Cave_Aquatic
AMNH 22645 Eurycea rathbuni 7.242 10.727 1.083 5.569 11.492 1.169 1.576 26.419 38.502 0.528 Cave_Aquatic
AMNH 22646 Eurycea rathbuni 8.963 14.441 1.592 7.132 14.458 1.799 1.591 35.454 46.041 0.419 Cave_Aquatic
AMNH 22647 Eurycea rathbuni 10.968 18.098 1.271 9.007 12.671 1.28 0.661 38.891 43.941 0.619 Cave_Aquatic
AMNH 51178 Eurycea rathbuni 10.868 17.498 1.548 8.488 16.589 1.427 0.944 39.659 47.826 Cave_Aquatic
AMNH 58632 Eurycea rathbuni 9.435 11.762 0.6 6.466 14.029 0.654 1.396 19.612 38.21 0.257 Cave_Aquatic
AMNH 58633 Eurycea rathbuni 7.66 12.89 0.786 5.515 15.126 1.122 0.928 30.47 39.804 0.464 Cave_Aquatic
AMNH 58634 Eurycea rathbuni 9.131 13.541 0.92 5.852 13.763 0.637 1.014 29.91 35.948 0.344 Cave_Aquatic
AMNH 62119 Eurycea rathbuni 8.709 12.498 1.096 5.096 16.424 1.462 0.724 33.19 46.13 0.221 Cave_Aquatic 52
AMNH 155730 Eurycea rathbuni 10.664 18.214 1.385 9.056 18.058 1.368 0.944 40.94 41.93 Cave_Aquatic
NMNH 37051 Eurycea spelaea 9.574 12.189 2.591 9.017 16.097 3.098 1.72 52.605 61.962 Cave_Terrestrial
NMNH 37052 Eurycea spelaea 9.625 10.293 2.116 8.231 9.96 2.682 1.687 50.168 56.719 Cave_Terrestrial
NMNH 37053 Eurycea spelaea 6.453 6.154 1.587 7.301 8.96 1.541 1.02 32.217 37.626 0.951 Cave_Terrestrial
NMNH 38787 Eurycea spelaea 9.179 10.222 2.434 10.263 11.7 2.648 1.74 52.528 55.343 1.729 Cave_Terrestrial
NMNH 38788 Eurycea spelaea 9.66 9.351 2.696 9.217 15.377 2.141 1.861 62.835 59.135 1.83 Cave_Terrestrial
NMNH 38789 Eurycea spelaea 7.831 10.05 1.966 8.129 9.351 2.558 1.413 58.32 47.7 1.474 Cave_Terrestrial
NMNH 54326 Eurycea spelaea 9.686 12.845 2.832 9.181 11.203 3.326 2.12 50.602 61.288 1.359 Cave_Terrestrial
NMNH 57327 Eurycea spelaea 8.797 11.938 2.456 9.19 12.274 3.03 1.83 56.315 61.181 1.082 Cave_Terrestrial
NMNH 57332 Eurycea spelaea 7.277 7.659 1.93 6.726 9.963 2.455 1.31 47.23 50.025 0.821 Cave_Terrestrial
NMNH 153780 Eurycea tridentifera 7.174 10.33 0.798 4.793 10.786 1.067 1.797 23.801 22.868 0.453 Cave_Aquatic
NMNH 153781 Eurycea tridentifera 6.16 6.947 0.54 3.555 7.417 0.904 1.022 27.779 29.677 0.602 Cave_Aquatic
NMNH 153782 Eurycea tridentifera 5.818 6.725 0.476 3.271 6.128 0.846 0.861 19.396 23.487 0.598 Cave_Aquatic
NMNH 153783 Eurycea tridentifera 4.693 5.917 0.558 2.983 7.277 0.899 0.933 21.884 22.235 1.232 Cave_Aquatic
NMNH 153784 Eurycea tridentifera 4.405 6.261 0.511 2.918 6.582 0.603 1.1 18.524 21.521 0.388 Cave_Aquatic
NMNH 153785 Eurycea tridentifera 4.212 5.359 0.482 2.373 6.397 0.495 0.672 18.813 18.997 0.43 Cave_Aquatic
AMNH 62060 Eurycea tynerensis 3.272 3.587 0.579 2.901 4.211 0.858 0.384 24.989 29.724 0.842 Non-cave_Aquatic
AMNH 62061 Eurycea tynerensis 3.972 5.091 0.717 3.054 5.646 0.708 0.745 26.345 28.529 0.892 Non-cave_Aquatic
AMNH 182184 Eurycea tynerensis 3.194 3.518 0.576 3.44 4.149 0.508 0.323 28.292 29.615 1.073 Non-cave_Aquatic
AMNH 80090 Eurycea wallacei 3.289 4.603 0.386 1.848 5.433 0.394 0.449 13.785 19.953 0.135 Cave_Aquatic
AMNH 80091 Eurycea wallacei 4.449 6.423 0.506 2.566 6.109 0.647 0.418 17.967 23.595 0.263 Cave_Aquatic
AMNH 172387 Eurycea wallacei 3.649 4.355 0.588 2.55 4.558 0.979 0.415 13.494 18.467 0.201 Cave_Aquatic
AMNH 182197 Eurycea wallacei 3.494 5.107 0.396 2.437 4.743 0.521 0.635 13.077 17.619 0.415 Cave_Aquatic
AMNH 99654 Eurycea wilderae 6.685 8.435 1.52 5.64 9.61 1.811 1.225 51.368 39.189 1.656 Non-cave_Terrestrial
AMNH 99655 Eurycea wilderae 6.076 8.215 1.276 5.57 9.741 1.645 1.26 53.578 40.35 1.84 Non-cave_Terrestrial
AMNH 99657 Eurycea wilderae 6.238 7.245 1.514 5.413 7.949 1.674 1.747 42.287 36.512 1.832 Non-cave_Terrestrial
AMNH 99658 Eurycea wilderae 5.77 7.604 1.446 4.929 8.422 1.876 1.226 50.049 38.119 1.681 Non-cave_Terrestrial
AMNH 99659 Eurycea wilderae 6.674 7.939 1.486 6.059 9.32 2.046 0.986 55.488 44.332 1.889 Non-cave_Terrestrial 53
AMNH 99660 Eurycea wilderae 5.948 8.373 1.423 5.155 10.014 1.602 1.269 53.004 40.795 1.831 Non-cave_Terrestrial
AMNH 115827 Eurycea wilderae 4.737 6.677 1.064 5.266 10.054 1.49 1.282 50.934 37.154 1.28 Non-cave_Terrestrial
AMNH 115829 Eurycea wilderae 5.153 7.902 1.122 5.379 10.438 1.456 1.369 58.034 39.118 1.258 Non-cave_Terrestrial
AMNH 115830 Eurycea wilderae 5.653 8.4 1.168 6.114 11.581 1.726 1.581 65.387 42.22 1.501 Non-cave_Terrestrial
AMNH 115833 Eurycea wilderae 6.095 7.743 1.26 6.179 10.168 1.925 1.335 58.364 42.062 1.428 Non-cave_Terrestrial
AMNH 115834 Eurycea wilderae 5.487 8.646 1.318 5.864 10.609 1.698 1.224 56.287 39.632 1.634 Non-cave_Terrestrial
AMNH 115835 Eurycea wilderae 5.054 8.263 1.179 5.253 10.484 1.56 1.329 55.29 36.557 1.301 Non-cave_Terrestrial
AMNH 115836 Eurycea wilderae 4.917 6.858 1.296 4.718 10.181 1.461 1.131 50.933 35.509 1.326 Non-cave_Terrestrial
AMNH 127036 Eurycea wilderae 4.34 6.101 0.957 4.199 6.196 1.233 0.845 38.961 32.577 1.283 Non-cave_Terrestrial
AMNH 127039 Eurycea wilderae 4.167 5.667 0.895 4.281 7.187 1.283 0.885 38.919 32.419 1.264 Non-cave_Terrestrial
AMNH 155790 Eurycea wilderae 4.81 7.12 1.019 4.852 9.405 1.196 0.863 30.622 32.061 1.218 Non-cave_Terrestrial
AMNH 155791 Eurycea wilderae 4.677 7.344 0.95 4.502 8.563 1.416 1.084 37.525 43.436 1.568 Non-cave_Terrestrial
AMNH 171738 Eurycea wilderae 5.59 7.081 1.242 5.084 8.506 1.307 1.323 26.146 44.497 1.663 Non-cave_Terrestrial
AMNH 171739 Eurycea wilderae 4.971 5.148 1.103 4.63 6.714 1.226 0.564 38.34 37.97 1.324 Non-cave_Terrestrial
AMNH 172392 Eurycea wilderae 5.701 9.065 1.164 6.381 8.876 1.437 1.237 33.108 39.246 1.772 Non-cave_Terrestrial
AMNH 95353 Gyrinophilus porphyriticus 9.207 8.174 2.01 10.051 8.367 2.447 1.54 40.395 55.099 1.274 Cave_Terrestrial
AMNH 137525 Gyrinophilus porphyriticus 13.079 15.153 3.378 12.838 16.849 3.367 1.609 41.484 93.658 3.669 Cave_Terrestrial
AMNH 137533 Gyrinophilus porphyriticus 10.641 9.615 2.438 9.989 10.624 2.641 1.263 31 62.48 2.006 Cave_Terrestrial
AMNH 137538 Gyrinophilus porphyriticus 12.043 9.944 2.701 11.31 14.977 3.082 1.758 42.143 70.606 2.928 Cave_Terrestrial
AMNH 137590 Gyrinophilus porphyriticus 10.455 9.398 2.608 9.724 9.797 2.481 1.136 40.265 57.712 1.254 Cave_Terrestrial
AMNH 137591 Gyrinophilus porphyriticus 10.137 9.323 2.077 10.24 11.572 2.71 1.124 39.248 56.742 1.482 Cave_Terrestrial
AMNH 137593 Gyrinophilus porphyriticus 9.552 9.827 2.517 10.382 9.118 2.533 0.998 42.651 58.975 1.427 Cave_Terrestrial
AMNH 137631 Gyrinophilus porphyriticus 12.852 11.205 3.073 11.496 16.193 3.885 2.194 45.029 78.49 2.401 Cave_Terrestrial
AMNH 137632 Gyrinophilus porphyriticus 10.887 13.713 2.612 10.309 15.807 3.199 1.705 38.977 75.936 2.353 Cave_Terrestrial
AMNH 137653 Gyrinophilus porphyriticus 17.023 17.691 4.719 18.167 21.134 5.143 2.991 72.221 119.842 3.954 Cave_Terrestrial
AMNH 137654 Gyrinophilus porphyriticus 16.159 18.13 3.91 17.058 21.519 5.219 2.795 76.475 113.408 3.256 Cave_Terrestrial
AMNH 157716 Gyrinophilus porphyriticus 13.842 16.283 3.578 17.532 19.846 3.698 2.948 87.135 85.935 2.639 Cave_Terrestrial
AMNH 169622 Gyrinophilus porphyriticus 24.362 11.092 3.201 12.077 13.516 3.345 2.658 62.303 77.954 2.542 Cave_Terrestrial 54
AMNH 170907 Gyrinophilus porphyriticus 8.679 9.963 2.313 9.908 13.957 3.683 2.948 46.11 68.947 3.182 Cave_Terrestrial
AMNH 182194 Gyrinophilus porphyriticus 10.777 12.333 2.01 10.443 18.434 2.787 1.361 55.287 77.817 2.469 Cave_Terrestrial
AMNH 190416 Gyrinophilus porphyriticus 14.652 12.795 3.46 15.776 18.803 4.277 4.27 45.85 106.782 2.717 Cave_Terrestrial
NMNH 285091 Hydromantes brunus 5.902 10.953 1.13 4.501 10.857 1.397 1.667 22.557 31.64 2.593 Non-cave_Terrestrial
NMNH 285093 Hydromantes brunus 5.345 6.743 0.938 5.22 6.994 1.39 1.205 17.264 23.634 1.928 Non-cave_Terrestrial
NMNH 285094 Hydromantes brunus 5.501 6.865 0.915 5.108 6.84 1.193 1.401 17.711 22.853 1.992 Non-cave_Terrestrial
NMNH 285097 Hydromantes brunus 4.432 5.9 0.667 3.315 6.102 1.494 0.985 15.192 19.85 1.761 Non-cave_Terrestrial
NMNH 321295 Hydromantes brunus 11.258 14.574 2.459 9.039 16.23 3.095 2.713 49.119 55.379 3.89 Non-cave_Terrestrial
NMNH 545723 Hydromantes brunus 7.445 9.051 1.204 5.223 10.578 1.519 1.53 21.744 30.875 2.522 Non-cave_Terrestrial
NMNH 58732 Hydromantes genei 7.205 9.129 1.66 5.728 9.547 1.535 1.031 22.551 33.639 2.787 Cave_Terrestrial
NMNH 58733 Hydromantes genei 4.408 4.93 0.908 3.277 5.795 0.951 0.546 11.897 18.331 1.295 Cave_Terrestrial
NMNH 93878 Hydromantes genei 10.681 15.11 2.485 10.033 18.979 4.292 2.169 50.849 58.325 3.44 Cave_Terrestrial
NMNH 93880 Hydromantes genei 10.462 16.68 2.365 8.494 22.756 2.449 2.651 49.441 59.173 3.625 Cave_Terrestrial
NMNH 100925 Hydromantes genei 10.072 19.464 2.04 7.631 17.777 2.499 2.627 61.391 41.524 3.185 Cave_Terrestrial
NMNH 100930 Hydromantes genei 10.379 15.683 1.836 8.09 18.378 2.704 2.574 48.151 53.573 3.297 Cave_Terrestrial
NMNH 100932 Hydromantes genei 9.114 16.772 1.778 8.006 14.462 2.667 1.904 37.051 52.767 2.97 Cave_Terrestrial
AMNH 23485 Hydromantes italicus 8.681 12.366 2.01 7.472 12.389 2.325 2.182 27.145 52.033 2.393 Cave_Terrestrial
AMNH 23645 Hydromantes italicus 9.448 13.842 1.837 7.541 10.626 2.605 1.75 32.263 50.672 2.329 Cave_Terrestrial
AMNH 24873 Hydromantes italicus 8.66 10.012 2.555 7.499 10.989 1.635 1.487 34.262 47.057 3.085 Cave_Terrestrial
AMNH 24874 Hydromantes italicus 9.674 15.047 2.506 8.552 13.641 2.21 2.905 41.339 53.212 2.561 Cave_Terrestrial
AMNH 34631 Hydromantes italicus 9.512 13.023 2.376 8.81 12.227 2.498 3.483 32.732 51.687 2.266 Cave_Terrestrial
AMNH 34632 Hydromantes italicus 8.82 15.057 2.071 7.072 16.549 1.992 1.547 37.729 54.756 2.119 Cave_Terrestrial
AMNH 34633 Hydromantes italicus 9.568 14.473 2.353 7.432 15.293 3.36 1.627 38.346 55.567 2.752 Cave_Terrestrial
AMNH 52692 Hydromantes italicus 9.453 11.126 1.826 7.426 11.926 2.327 1.757 34.411 49.702 2.522 Cave_Terrestrial
AMNH 54694 Hydromantes italicus 10.526 18.021 2.081 10.356 17.261 2.54 1.779 33.898 65.84 2.71 Cave_Terrestrial
AMNH 54695 Hydromantes italicus 10.329 14.518 2.081 8.642 14.167 3.415 1.593 38.649 59.431 2.369 Cave_Terrestrial
AMNH 54696 Hydromantes italicus 11.581 16.005 2.446 9.39 15.088 3.116 1.311 45.963 62.773 2.983 Cave_Terrestrial
AMNH 65109 Hydromantes italicus 9.903 13.373 2.091 7.359 14.133 2.975 1.325 35.788 58.492 2.507 Cave_Terrestrial 55
AMNH 65110 Hydromantes italicus 9.685 10.16 1.946 7.604 14.598 2.421 1.193 32.221 55.705 2.459 Cave_Terrestrial
AMNH 65111 Hydromantes italicus 9.043 14.669 1.826 6.554 15.533 2.071 1.375 39.774 59.958 2.668 Cave_Terrestrial
AMNH 65112 Hydromantes italicus 6.749 8.192 1.145 4.972 10.269 1.448 1.042 23.099 34.535 1.968 Cave_Terrestrial
AMNH 149539 Hydromantes italicus 8.539 10.594 1.441 6.36 12.542 1.761 1.643 28.134 51.758 2.327 Cave_Terrestrial
AMNH 149540 Hydromantes italicus 9.067 12.844 1.662 7.85 12.903 2.338 1.863 32.22 50.977 1.839 Cave_Terrestrial
AMNH 149541 Hydromantes italicus 10.479 14.251 1.94 8.703 14.55 2.22 1.45 42.806 58.289 2.422 Cave_Terrestrial
AMNH 149542 Hydromantes italicus 11.107 14.12 2.255 9.779 14.971 3.408 1.104 35.204 59.952 2.587 Cave_Terrestrial
AMNH 149543 Hydromantes italicus 9.818 13.192 2.059 8.985 13.341 2.82 1.473 40.591 56.153 2.3 Cave_Terrestrial
AMNH 53808 Hydromantes platycephalus 8.704 12.358 1.823 8.947 15.483 2.433 1.863 35.84 56.365 2.593 Non-cave_Terrestrial
AMNH 53809 Hydromantes platycephalus 8.541 11.476 1.554 7.626 12.031 1.659 0.984 21.746 42.696 2.359 Non-cave_Terrestrial
AMNH 53810 Hydromantes platycephalus 6.939 7.574 1.294 6.722 10.169 1.586 1.044 16.498 33.235 2.017 Non-cave_Terrestrial
AMNH 53811 Hydromantes platycephalus 6.556 7.825 1.1 5.754 9.765 1.828 1.083 20.158 32.934 1.497 Non-cave_Terrestrial
AMNH 53812 Hydromantes platycephalus 9.477 10.711 1.816 8.55 14.447 2.272 1.168 30.359 51.079 2.062 Non-cave_Terrestrial
56
CHAPTER 2
THE PHYLOGEOGRAPHY AND POPULATION GENETICS OF A CAVE- DWELLING SALAMANDER, EURYCEA LUCIFUGA
Abstract The evolutionary history and ecology of cave-dwelling species has been driven historically by studies of morphologically adapted cave-restricted species. As a result, our understanding of troglophiles’ evolutionary history and ecology is limited to a few studies, which present differing accounts of troglophiles’ relationship with the cave habitat, and its impact on population dynamics. We used phylogenetics, demographical statistics, and population genetic methods to study lineage divergence and population structure in the cave salamander Eurycea lucifuga across its range, as well as inferring dates of divergence among major clades. Results of phylogenetic and demographic analyses indicate that there are three main lineages within E. lucifuga corresponding to the Western, central, and Eastern regions of the range. Divergence among these major regions dates to the Pliocene and Pleistocene, with evidence of recent expansion into present-day localities. Population genetic analyses largely corroborate these results, but indicate isolation within the central clade between populations located in Indiana and those in Kentucky and Tennessee. Genetic diversity as well as a Bayesian analysis of divergence scenario likelihoods suggest that populations may have expanded from the
Southern central areas of its range in two steps: first an expansion to the West and then another to the East. Correlation of divergence among regional lineages in Eurycea lucifuga with the dramatic climatic and geological shifts of the Pliocene and Pleistocene is similar to what we see in other cave-dwelling species. 57
Introduction:
Examining populations’ evolutionary histories in the context of their spatial distribution often gives us a better understanding about the formation of lineages and species, which can then be used in studies of adaptation and morphological evolution. A useful area of study for questions of this nature is the existence of cave-dwelling species, and their divergence from surface-dwelling relatives. The vast majority of research on cave systems has focused on troglobites (cave-restricted morphologically adapted species), which are characterized by a suite of acquired traits known as troglomorphy.
These morphological and physiological traits are thought to act both as a mechanism behind subterranean speciation (Barr et al. 1968; Culver 1982) and as a constraint upon dispersal and gene flow in troglobites (Porter 2007). For this reason, troglobitic cave- dwellers are often characterized by endemism and limited range (Christman et al. 2005;
Culver & Holsinger 1992); for example, the majority of US troglobites are endemic to a single county (Culver et al. 2000).
Although a great deal of attention is paid to troglobitic and/or troglomorphic species in the literature, there are other classes of cave-dwelling species. Of particular interest here is a class of cave-dwellers known as troglophiles. These species maintain permanent cave populations, but vary in their restrictedness to the cave habitat (Sket
2008). For some taxa, not being restricted to the cave habitat allows for greater dispersal in troglophilic populations than in troglobitic populations. Caccone (1985) studied several species of cave arthropods of varying cave restrictedness and found that while the amount of gene flow between populations and population structure varied tremendously among all species, and were highly dependent on both the characteristics of the cave and 58 surface environments as well as innate characteristics of the species, troglobites tended to have lower levels of gene flow among caves than troglophiles (Caccone 1985). In contrast, populations of the troglophilic isopod Androniscus dentiger exhibited surprisingly high differentiation with no detectable inter-population gene flow (Gentile &
Sbordoni 1998). Similarly, very low levels of gene flow and high population structure have been described in troglophilic Sardinian cave salamanders (genus Hydromantes), likely due to the inhospitable surface environments that discourage dispersal outside of caves (Chiari et al. 2012). These studies suggest that, depending on environmental and species characteristics, troglophiles may experience restriction to caves in the same way that troglobites do, though perhaps for different reasons. However, the relative lack of attention paid to troglophilic species prevents us from understanding how inhabiting the cave environment will impact the population and demographic dynamics of a species where cave association is not mandated by morphological restrictions, as it is in troglobitic species.
Study system
We studied the phylogeographic history and population dynamics of a troglophilic salamander in the genus Eurycea (Plethodontidae). Eurycea is particularly notable for its abundance of troglobitic species. Though surface-dwelling is the ancestral state in this genus, cave-dwelling, and troglomorphy, evolved independently at least five times across the genus (Bonett et al. 2014). Eurycea lucifuga is a troglophilic salamander commonly found in limestone caves throughout the Southeast and Midwest United States (Figure 1).
Eurycea lucifuga has a multi-stage life cycle, in which individuals are restricted to aquatic cave habitats in the egg and larval stages before fully metamorphosing into 59 brightly colored adults (Petranka 1998). There are anecdotal reports of E. lucifuga being in the forest matrix surrounding caves (Hutchison 1958; Petranka 1998), though cave habitats are by far the most common environment for E. lucifuga, and individuals are often found large distances (>1000m) beyond cave entrances (H.E. pers. obs.).
Despite the frequency with which Eurycea lucifuga is found across its wide range and its occupancy of a charismatic and conservationally important niche, very little is known about its evolutionary history or for how long it has inhabited caves.
Understanding these important attributes is a critical component of examining other aspects of its ecology and life history. We investigated the phylogeography of Eurycea lucifuga using phylogenetic, population genetic, and demographic statistical methods in order to better understand its evolutionary history. First, we used a tree-building approach to reconstruct the phylogeographic history of populations across the range of E. lucifuga and estimated the timing of divergence between populations in different geographic regions. Second, we used a classical population genetics approach to assess how genetic variation is structured across the range, and to infer a reconstruction of historic lineage splitting.
Methods:
Data collection
Between 2012 and 2013 we collected tail clips from between one and nine individuals from populations of Eurycea lucifuga throughout its range (Figure 1). A population is defined here as a group of individuals collected from inside an entrance to a cave. This definition can be problematic in cases where cave entrances are part of a larger cave system; however, to the best of our knowledge the majority of cave entrances 60 in this study are unconnected from other entrances. Animals were released at their capture site following tissue extraction.
Tissue samples were stored in 95% ethanol at room temperature at the University of Virginia (Charlottesville, VA). We extracted total genomic DNA from each sample using either a phenol-chloroform method (Sambrook et al. 1989) or Chelex resin, using a
5-10% slurry of Chelex and an incubation time of 180 minutes at 95°C. We then amplified three gene fragments commonly used in salamander systematics using polymerase chain reaction (PCR): two mitochondrial loci, NADH dehydrogenase 2
(ND2) and cytochrome b (cytb), and one nuclear locus, proopiomelanocortin (POMC).
Primer information and thermocycling conditions are shown in Table S1 (Supporting information). We cleaned the PCR products using ExoSAP-IT (Affymetrix) and then sequenced them in both directions using Sanger sequencing on an ABI 3730xl at
GENEWIZ (South Plainfield, NJ). We used the software program Geneious v 6.1.6
(Biomatters, Aukland, New Zealand) to edit the resulting sequence, then used the
Geneious alignment option in Geneious to create multiple alignments for each locus.
Alignments were unambiguous, and there were no gaps in the sequence, nor heterozygous sites in the nuclear locus. We included sequence data from E. longicauda and Pseudotriton ruber as outgroups. We collapsed identical haplotypes using the web- based software program ALTER (Glez-Peña et al. 2010). All sequences are deposited in
GenBank (For collection records including accession numbers see Table S2, Supporting information). We used Xia’s Index of Sequence Saturation in the software package
DAMBE (Xia & Lemey 2009; Xia et al. 2003) to test whether each locus was phylogenetically informative. 61
We also genotyped each individual at 19 microsatellite loci designed specifically for E. lucifuga (Appendix 1). Loci were amplified in multiplexed reactions using PCR, and products were sent to the DNA Analysis Facility on Science Hill at Yale University
(New Haven, CT) for fragment analysis using a 3730xl 96-Capillary Genetic Analyzer, with the DS-33 dye set. Finally, we used GeneMarker software (SoftGenetics LLC, State
College, PA) to call alleles at each marker locus for each individual.
Phylogenetic analyses
We analyzed the sequence data from cytb, ND2, and POMC individually and as a combined, partitioned dataset under maximum likelihood (ML) and Bayesian inference
(BI) using RAxML 7.2.6 (Stamatakis 2006) and MrBayes 3.1.2 (Ronquist et al. 2012).
Individuals with missing data were excluded from any reconstructions where the loci were included in a combined analysis. Appropriate models of sequence evolution were inferred for each locus using both the Akaike Information Criterion (AIC), Bayesian
Information Criterion (BIC), and Decision Theory Performance-Based Selection (DT) in the program jModelTest v 2.1.4 (Posada 2008). We first performed ML phylogenetic analysis using the ML + thorough bootstrap analysis option with 2 independent runs and
1000 nonparametric bootstraps. We then used Bayes Factors implemented in MrBayes to compare the heuristic mean likelihood of two BI analyses, in which one model specified partitioning among codons and one model specified a simpler partitioning scheme with only partitioning among the gene loci. This comparison indicated that the simpler partitioning model was more likely (individual codon partitioning: likelihood = -6000.91; combined codon partitioning: likelihood = -5788.31). Upon incorporating the correct partitioning model as well as appropriate substitution rates, each MrBayes analysis was 62 run for 50,000,000 generations and trees were sampled every 2,500 generations, with 3 heated chains and one unheated chain. After analyzing Bayesian results for convergence as well as comparing individual searches using Tracer (Drummond et al. 2012), the first
20% of the generations were discarded as a burn-in period. We used TreeAnnotator
(Drummond et al. 2012) to sample the trees and infer a single tree for each analysis using
Maximum Clade Credibility. We visualized trees using the program FigTree v 1.4.2
(Drummond et al. 2012), including ML bootstrapping and BI posterior probability information on the resulting BI tree at each well-supported major clade.
We also used the program BEAST v1.8.1 (Drummond et al. 2012) to recreate a species tree using the three partitioned sequence datasets. We used a stepping-stone analysis comparing the use of a relaxed molecular clock with a strict clock model implemented in MrBayes for 50 steps, each with 19,500 generations, which indicated that the strict clock model had a higher mean marginal likelihood (strict: Marginal likelihood
(ln) = -6288.50, relaxed: Marginal likelihood (ln) = -6334.81). Accordingly, we used a published rate to calibrate the clock for cytb and nd2 (0.0062 substitutions/my and 0.0037 substitutions/my respectively; (Mueller 2006)), and estimated the rates of POMC with reference to those rates. We ran the analysis two times independently for 500 million generations, sampling trees every 20,000 generations under a Yule model with piecewise linear population size and a constant root. After using Tracer (Drummond et al. 2012) to compare the independent runs and test for convergence we discarded the first 20% of the resulting trees as a burn-in, and inferred a single tree with TreeAnnotator (Drummond et al. 2012). This tree was visualized using Figtree v 1.4.2 (Drummond et al. 2012). We also performed mismatch analysis and calculated Tajima’s D and Fu’s F statistics in 63
Arlequin v.305 (Excoffier & Lischer 2010) to test for evidence of recent population expansions or bottlenecks for each of the major divergence events predicted by our phylogenetic analyses as a comparison with divergence times predicted by BEAST.
Using methods outlined by Holsinger (2008), we used the estimates of tau produced by mismatch analysis to calculate the time of the predicted expansion for each geographic group using the published rates for both cytb and nd2 (Mueller 2006).
Population genetic analyses
We performed a cluster analysis of the microsatellite genotype data using the program STRUCTURE v.2.3.4 (Pritchard et al. 2000). Following the methods of Coulon et al. (2008), we first ran the analysis with K=2, then analyzed each partition separately using a range of K=1-10. The most appropriate number of clusters was found by analyzing the results using the Evanno method (Evanno et al. 2005) in the web based software Structure Harvester (Earl 2012). We then visualized the clustering in distruct
(Rosenberg 2003). To test whether the pattern of variation could be explained by isolation by distance, we used the software program Genepop (Raymond & Rousset
1995; Rousset 2008) to examine the correlation between genetic and geographic distances with Mantel tests. We performed a separate analysis within each of the main clusters found by STRUCTURE using 1000 permutations of the Mantel tests. We examined genetic diversity within each region by comparing estimations of heterozygosity and the inbreeding coefficients directly. To estimate how variation is partitioned among- and within-groups and populations, and to estimate Wright’s F statistics, we used Analysis of Molecular Variance (AMOVA) in the program GenoDive
(Meirmans & Van Tienderen 2004). Finally, to compare the topologies predicted by the 64
Likelihood and Bayesian trees with those predicted by the BEAST species reconstruction and the microsatellite structuring data, we used Approximate Bayesian Computation in the program DIYABC (Cornuet et al. 2014) to infer the likelihood of different divergence scenarios. We tested four scenarios (Figure 2), one in which a radiation suggests an equal age for each major clade, and three more with each combination of divergences.
We kept the default priors for each analysis, and ran the simulation for 400,000 iterations.
Following the simulation, we performed model selection using both a direct estimation of the most likely scenario with 500 iterations as well as a logistical regression using 4000 iterations. Model checking was performed to compare axes of variation within the summary statistics produced from the simulation to those predicted by the posterior predicted distribution of the best-fitting model.
Results
Fragments of cytb, ND2 and POMC (1582bp total) from 93 individuals from 27 populations were included in our phylogenetic reconstructions and demographic statistical analyses. jModelTest (Posada 2008) predicted the most appropriate model of evolution for each locus to be TPM2uf+I+G (cytb), TIM3+I (ND2), and TIM2+G
(POMC). Since these models are not available options in most phylogenetic reconstruction programs, we used the slightly overparameterized models GTR+I+G,
GTR+I, and HKY+G, respectively during the analyses in MrBayes and RaxML, and
BEAST. None of the loci exhibited significant saturation (cytb: T=49.061, Df=463, p<0.0001; nd2: T=62.6957, Df=549, p<0.0001; POMC: T=122.5041, Df=465, p<0.0001), indicating that our sequence data contain useful phylogenetic information. 65
We genotyped 233 individuals from 53 populations at 19 microsatellite loci.
Summary statistics from the genotyped dataset can be found in Appendix A. The number of alleles per locus ranged from 2-7. With very few exceptions, all loci were in Hardy-
Weinberg equilibrium within populations (Appendix A, Table 4).
Phylogeographic analyses
The ML and Bayesian concatenated trees (Figure 3) both feature two well- supported major clades organized by geographic region; the topologies are very similar with only minor rearrangements within major clades. The central clade contains populations in western Tennessee, Kentucky, and Indiana, and the Eastern/Western clade contains the populations within Virginia, West Virginia, Eastern Tennessee, Oklahoma, and Missouri. The Western populations form a monophyletic clade within the Eastern populations. Haplotypes were shared among populations within the major geographic regions; however, support at the tips was very low. Relationships among the clades differ from simple geographic expectations: the Western and Eastern regions are more closely related to each other than either is to the central clade. Gene trees of individual loci (Figs
S1-3, Supporting Information) generally reflect these relationships, although the nuclear locus POMC did not have a large influence on the topology, and as a result the combined trees generally represent the mitochondrial history of the species. However, the topology of the tree produced using species tree methods in BEAST indicate that all three geographic regions form distinct clades, and that the central and Eastern populations are more closely related to each other than they are to the Western populations (Figure 4).
While branching leading to the Western clade is well-supported, with posterior probabilities of 1, the relationships among populations in the Eastern and central clades 66 are not as well supported. This is reflective of the fact that BEAST occasionally reconstructed a different topology for the tree, similar to that of the MrBayes and RaxML predictions with the Eastern and Western clades forming a monophyly and the central clade less closely related. Though the inconsistencies between the two reconstruction methods indicate some ambiguity in the DNA sequence data, the improved performance of species tree methods in general (Degnan & Rosenberg 2006; Heled & Drummond
2010; Liu & Pearl 2007) suggests to us that the BEAST reconstruction is the more accurate scenario. This is supported by results of the divergence dating and population genetic analyses we performed.
Divergence dating
The chronogram produced in BEAST (Figure 4) dates divergence from Eurycea longicauda at approximately 4 million years ago (mya), and between the two major clades (Eastern/Western and Central regions) at almost 2 mya. Divergence within each region was dated at more recently than a million years ago. Estimates of Tajima’s D and
Fu’s F statistics corroborate this evidence of recent population expansion within each major clade, with significantly negative values of Fu’s F, Tajima’s D, or both parameters for each expansion event (Table 1). Mismatch analysis did not indicate a departure from the null hypothesis of recent expansion in any group either with a significant SSD value or significant values for Harpending’s raggedness index. Using Holsinger’s (2008) methods for inferring timing of divergence from an estimate of tau, we found that expansion of the Western populations following divergence from the Central populations may have occurred in the range of one to three million years ago, the Eastern clade approximately 1-2.5 mya, and the Central populations experiencing expansion much 67 more recently, on the order of tens of thousands of years ago (Table 1). This timing, and the topology of the species tree reconstructed using BEAST are supported by the results of an analysis comparing different divergence scenarios using DIYABC, which predicted that the most likely scenario includes a primary splitting of the Western populations, followed by a subsequent divergence between the central and Eastern populations (Table
2, Figure 2).
Population genetic analyses
The STRUCTURE analysis indicates that the model with the highest likelihood is
K=4 with a mean likelihood of -4198.9. Assignment to these genetic clusters partitions the samples into clear geographic groups: Western, Indiana, South-central (Kentucky and
Tennessee), and Eastern populations (Figure 5). We did not find significant evidence of isolation by distance within the Western or Eastern populations; however, correlations between genetic and geographic distances were significant among populations in the northern Central region (r=0.05, p=0.023) and in the Southern central region (r=0.07, p=0.043). When compared directly, the central regions contain significantly more heterozygosity than the Western or Eastern regions (Table 3), so there may have been greater power to detect subtle patterns of isolation by distance. AMOVA results indicate significant population structure among sampling localities within the Eastern, Central and
Western regions (FSC=0.093, p=0.001) and strong population genetic divergence among regions (FCT=0.331, p=0.001). 33% of genetic variance is partitioned among regions, and
6% among localities within each region. A substantial amount of variation is partitioned among individuals within localities (13%, FIS=0.208, p=0.001) (Table 4).
68
Discussion
Results of the phylogenetic analyses and demographic statistics suggest a pattern of ancient lineage divergences within Eurycea lucifuga, followed by more recent population expansion within each region. The timing of divergence among the clades estimated in BEAST (1.7-2.7mya) is similar to lineage splitting in other cave-dwelling species (Finston et al. 2007; Lefébure et al. 2006; Leys et al. 2003; Strecker et al. 2004;
Zakšek et al. 2007), which lends support to Wessel’s (Wessel et al. 2007) finding that the age of a cave-dwelling lineage is not correlated with its morphological adaptation to the caves. This age places divergence within E. lucifuga during the late Pliocene or early
Pleistocene, in the midst of climatic and geological change, but well before the last glacial advance. The sharing of haplotypes among localities in concert with the results of Tajima’s D, Fu’s F, and mismatch analyses suggest that within each major region expansion has been recent; however, our estimations of the timing of that expansion indicate that it was well before the retreat of the last glacial ice sheet at the end of the
Pleistocene. As the last 700,000 years brought about particularly extreme climatic oscillations (Hewitt 1996; Webb & Bartlein 1992), the coincidence of expansion with that time suggests that the recent expansion we estimate was a result of that period of global change. It is probable that oscillating climatic conditions caused a series of expansions and contractions in the range of E. lucifuga; however, signatures of subsequent expansions would be overwhelmed by the initial expansion (Rogers 1995).
Population genetic analyses reveal that the majority of genetic variation across the range of Eurycea lucifuga is partitioned among the major geographic regions, and populations structure within each region is quite low (RST=0.038). This low structure 69 particularly contrasts with closely related cave-restricted species (E. pterophila/E. nana/E. neotenes: FST=0.249-0.922; (Lucas et al. 2009)) and with other Plethodontids studied using protein sequences, in which estimates of FST ranged from 0.13 (Plethodon cinereus) to 0.80 (Plethodon dorsalis dorsalis) (Larson et al. 1984). However, population structure is similar to estimates seen at smaller scales in P. cinereus
(FST=0.019; Cabe et al. (2007)) and Ambystoma maculatum (FST=0.041; Purrenhage et al.
(2009)). While low population structure could be a signature of the recent expansion predicted by our demographic statistical analysis, it also suggests a pattern of persistent gene flow among populations. Interestingly, there are distinct differences in population genetic structure and diversity among the main geographic clades of E. lucifuga, with the central populations exhibiting both greater observed and expected heterozygosity and more population structure than the Western and Eastern populations. Although we expect that central populations will have greater heterozygosity than peripheral populations, this is usually accompanied by higher population structure in peripheral populations (Eckert et al. 2008). We speculate that in this case, because movement between caves is often dependent on the characteristics of the surface environment
(Caccone 1985; Chiari et al. 2012), it is possible that the variations in climate and landscape in the different regions impacts this species’ dispersal ability, causing variation in population structure.
In reconstructing the phylogeographic history of a species it is important to keep in mind the geological and climatic processes that may have influenced a species’ present distribution. The divergence dates estimated using BEAST and demographic statistics indicate that Eurycea lucifuga’s population dynamics have mainly been influenced by the 70 late Pleiocene and early Pleistocene eras, which was a time of great climatic and geological change. The earth was shifting from a warming period during the Miocene
(~24-5mya) to a cooling period from the Pliocene (~5-2mya) until glacial retreat at the end of the Pleistocene (~0.01mya). Throughout the Pliocene and Pleistocene glacial advance and retreat oscillated a number of times, and increased rainfall in non-glaciated areas benefited the species that took refuge there (Levin 2009). At its most extensive time, the Laurentide ice cap reached as far South as forty degrees latitude (Stearn et al.
1979). Fluctuations in temperature and ice cover caused corresponding oscillations in species’ ranges, which often led to a series of bottleneck at range edges, resulting in reduced genetic diversity among those edge populations (Hewitt 1996). These expansions and contractions applied equally to warm-adapted species during the onset of glaciation, and cold-adapted species during inter-glacial periods (Hewitt 1996). During glaciation, many species retreated to unglaciated areas known as glacial refugia. Often the use of multiple glacial refugia was a cause of divergence among lineages, (Bossu et al. 2013; Phillips 1994; Zamudio & Savage 2003), and many species show signs of intraspecific divergence dating to periods of glaciation (Avise 2000; Kozak et al. 2006;
Niemiller et al. 2008), although pre-glaciation divergence has been estimated in other species (Burbrink et al. 2008; Hoffman & Blouin 2004; Jones et al. 2006). Refugia have been predicted in areas both East and West of the Mississippi River in the central basin, as well as in the Southern Appalachians (Bossu et al. 2013; Hoffman & Blouin 2004;
Rissler 2010; Taylor et al. 2007; Zamudio & Savage 2003).
Following glacial retreat, species’ dispersal from glacial refugia and expansion into the present-day ranges involved certain patterns repeated across taxa. Range 71 expansions led to a significant clustering of amphibian taxa in areas with very high species richness: primarily in the Appalachian Mountains and throughout Alabama
(Rissler 2010). Many taxa exhibit consistent phylogenetic breaks across five ‘barrier zones’ across the United States: between the Atlantic and Gulf coasts, the Appalachian
Mountains, and across the Appalachicola River (SE United States), the Tombigbee River
(Alabama), and the Mississippi River (Soltis et al. 2006). There are three alternative scenarios we could predict given these trends and the results of our analyses:
1) The clear divisions among geographic lineages across the range could indicate that each clade expanded from a different glacial refugium, located to the West of the
Mississippi River, East of the Mississippi in the central lowlands, and/or in the Southern
Appalachians. This would be a similar history to that of Rana pipiens (Hoffman &
Blouin 2004) or Ambystoma maculatum (Zamudio & Savage 2003).
2) Given the difference in genetic diversity and heterozygosity between the central clade and the peripheral clades, an alternative scenario is that of a single refugium in the central basin to the East of the Mississippi. Following expansion northward, the
Western and Eastern clades may have formed as a result of separate migration events, and their low diversity could then be attributed to founder effects. This would explain the inclusion of Eastern and Western alleles within the central populations.
3) It could be predicted that range expansion in Eurycea lucifuga was closely tied to geomorphological history throughout this time in addition to climatic fluctuations due to its cave inhabitancy. It has been suggested that the caves served as refugia for cold- adapted species during the warming of interglacial periods (Barr & Holsinger 1985), which could be the case in this species, particularly given the timing of expansion in each 72 region. While cave formations are particularly ancient in the Ozarks (ranging from
Cambrian to Mississippian- 530 to 330 mya) (Elliott & Ashley 2005), cave ages in the
Appalachians, specifically the Cumberland, ranged from 5.68-0.02mya (Anthony &
Granger 2007). Mammoth Cave was formed between 2.4-2.3mya, but experienced instability until around 0.7mya (Granger et al. 2001). If Eurycea lucifuga retreated into the cave habitat in order to avoid the warming climate of an interglacial period, it is plausible that this would have occurred earlier in the Western region where caves had been stable for millions of years, whereas establishment in caves in the central and
Eastern regions would not have occurred until those formations were structurally stable.
The results of species tree reconstructions, genetic clustering using
STRUCTURE, and results of DIYABC, in addition to the increased diversity within the central clade, suggest that the most likely scenario is that the Western clade diverged first, followed by a split between the central and Eastern populations. Sequential dispersal from the central region to the West and East would explain our observation that the central clade contains the most genetic diversity as well as the highest heterozygosity of the clades.
While we cannot distinguish among these hypotheses at present, our results indicate that Eurycea lucifuga is a species of early Pliocene origin, and that its distribution and ecology have been largely influenced by the climatic and geological events of the late Pliocene and early Pleistocene. While we may have expected expansion into present cave localities to have occurred recently, following the last glacial retreat, the timing of divergence and regional population expansion is similar to that of other cave species. However, evidence of recent gene flow among populations is 73 suggestive that the ecology of this species differs from most cave-dwelling taxa, which show limited gene flow. Incorporating this information with that of other systems will benefit our understanding of how species have changed with past earth events, and predict what their future evolutionary and ecological trajectories may be.
74
Acknowledgements
The authors would like to acknowledge the assistance of G. Moni, T. Loring, J.
Lewis, K. Dunlap, D. Everton, J. Pearson, S. Hammond, C. Evans, M. Futrell, A. Futrell,
W. Orndorff, R. Stewart, R. Toomey, R. Olson, B. Balfour, and many other caving enthusiasts and wildlife professionals at each sampling locality in the collection of samples. Collections were made with the permission of the Missouri Department, West
Virginia Division of Natural Resources, Kentucky Department of Fish and Wildlife,
United States Department of the Interior National Park Service Mammoth Cave National
Park, Indiana Department of Natural Resources Division of Fish and Wildlife, Oklahoma
Department of Wildlife Conservation, The Nature Conservancy (Mill Creek Cave),
Virginia Department of Game and Inland Fisheries, and the Virginia Department of
Conservation and Recreation. In addition, we would like to thank Dr. Robert Cox and Dr.
Edmund D. Brodie III for their comments during the writing process, and Dr. Peter D.
Fields for very helpful input at all stages of this project.
75
References
Anthony DM, Granger DE (2007) A new chronology for the age of Appalachian
erosional surfaces determined by cosmogenic nuclides in cave sediments.
Earth Surface Processes and Landforms 32, 874-887.
Avise JC (2000) Phylogeography: the history and formation of species Harvard
University Press, Cambridge, MA.
Barr T, Dobzhansky T, Hecht M, Steere W (1968) Cave ecology and the evolution of
troglobites. In: Evolutionary Biology, pp. 35-102. Appleton-Century-Crofts,
New York.
Barr T, Holsinger J (1985) Speciation in cave faunas. Annual Review of Ecology and
Systematics 16, 313-337.
Bonett RM, Steffen MA, Lambert SM, Wiens JJ, Chippindale PT (2014) Evolution of
paedomorphosis in plethodontid salamanders: ecological correlates and re-
evaluation of metamorphosis. Evolution 68, 466-482.
Bossu CM, Beaulieu JM, Ceas PA, Near TJ (2013) Explicit tests of palaeodrainage
connections of southeastern North America and the historical biogeography
of Orangethroat Darters (Percidae: Etheostoma: Ceasia). Molecular Ecology
22, 5397-5417.
Burbrink FT, Fontanella F, Alexander Pyron R, Guiher TJ, Jimenez C (2008)
Phylogeography across a continent: the evolutionary and demographic
history of the North American racer (Serpentes: Colubridae: Coluber
constrictor). Mol Phylogenet Evol 47, 274-288. 76
Cabe PR, Page RB, Hanlon TJ, et al. (2007) Fine-scale population differentiation and
gene flow in a terrestrial salamander (Plethodon cinereus) living in
continuous habitat. Heredity (Edinb) 98, 53-60.
Caccone A (1985) Gene Flow in Cave Arthropods: A Qualitative and Quantitative
Approach. Evolution 39, 1223-1235.
Chiari Y, van der Meijden A, Mucedda M, et al. (2012) Phylogeography of Sardinian
Cave Salamanders (Genus Hydromantes) Is Mainly Determined by
Geomorphology. PLoS ONE 7, e32332.
Christman MC, Culver DC, Madden MK, White D (2005) Patterns of endemism of the
eastern North American cave fauna. Journal of Biogeography 32, 1441-1452.
Cornuet J-M, Pudlo P, Veyssier J, et al. (2014) DIYABC v2.0: a software to make
approximate Bayesian computation inferences about population history
using single nucleotide polymorphism, DNA sequence and microsatellite
data. Bioinformatics.
Coulon A, Fitzpatrick JW, Bowman R, et al. (2008) Congruent population structure
inferred from dispersal behaviour and intensive genetic surveys of the
threatened Florida scrub-jay (Aphelocoma coerulescens). Mol Ecol 17, 1685-
1701.
Culver D (1982) Cave life: Evolution and Ecology Harvard University Press,
Cambridge, MA.
Culver D, Master L, Christman M (2000) Obligate cave fauna of the 48 contiguous
United States. Conservation Biology. 77
Culver DC, Holsinger JR (1992) How many species of troglobites are there. National
Speleological Society Bulletin 54, 59-80.
Degnan JH, Rosenberg NA (2006) Discordance of Species Trees with Their Most
Likely Gene Trees. PLoS Genet 2, e68.
Drummond AJ, Suchard M, Xie D, Rambaut A (2012) Bayesian phylogenetics with
BEAUti and the BEAST 1.7. Molecular Biology and Evolution 29, 1969-1973.
Earl DA (2012) STRUCTURE HARVESTER: a website and program for visualizing
STRUCTURE output and implementing the Evanno method. Conservation
Genetics Resources 4, 359-361.
Eckert CG, Samis KE, Lougheed SC (2008) Genetic variation across species’
geographical ranges: the central–marginal hypothesis and beyond. Molecular
Ecology 17, 1170-1188.
Elliott WR, Ashley DC (2005) Caves and karst. Nelson, Paul, The Terrestrial Natural
Communities of Missouri, third ed. Missouri Natural Areas Committee, 474-491.
Evanno G, Regnaut S, Goudet J (2005) Detecting the number of clusters of
individuals using the software STRUCTURE: a simulation study. Molecular
Ecology 14, 2611-2620.
Excoffier L, Lischer HEL (2010) Arlequin suite ver 3.5: A new series of programs to
erform population genetics analyses under Linux and Windows. Molecular
ecology resources 10, 564-567.
Finston TL, Johnson MS, Humphreys WF, Eberhard SM, Halse SA (2007) Cryptic
speciation in two widespread subterranean amphipod genera reflects 78
historical drainage patterns in an ancient landscape. Molecular Ecology 16,
355-365.
Gentile G, Sbordoni V (1998) Indirect Methods to Estimate Gene Flow in Cave and
Surface Populations of Androniscus dentiger (Isopoda: Oniscidea). Evolution
52, 432-442.
Glez-Peña D, Gómez-Blanco D, Reboiro-Jato M, Fdez-Riverola F, Posada D (2010)
ALTER: program-oriented conversion of DNA and protein alignments. Nucleic
Acids Research 38, W14-W18.
Granger DE, Fabel D, Palmer AN (2001) Pliocene−Pleistocene incision of the Green
River, Kentucky, determined from radioactive decay of cosmogenic 26Al and
10Be in Mammoth Cave sediments. Geological Society of America Bulletin
113, 825-836.
Heled J, Drummond AJ (2010) Bayesian Inference of Species Trees from Multilocus
Data. Molecular Biology and Evolution 27, 570-580.
Hewitt GM (1996) Some genetic consequences of ice ages, and their role in
divergence and speciation. Biological Journal of the Linnean Society 58, 247-
276.
Hoffman EA, Blouin MS (2004) Evolutionary history of the Northern Leopard Frog:
reconstruction of phylogeny, phylogeography, and historical changes in
population demography from mitochondrial DNA. Evolution 58, 145-159.
Holsinger K (2008) Population Genetics. University of Connecticut.
http://darwin.eeb.uconn.edu/eeb348/lecturenotes/mismatch-
analysis/node3.html 79
Hutchison V (1958) The Distribution and Ecology of the Cave Salamander, Eurycea
lucifuga. Ecological Monographs 28, 1-20.
Jones MT, Voss SR, Ptacek MB, Weisrock DW, Tonkyn DW (2006) River drainages
and phylogeography: an evolutionary significant lineage of shovel-nosed
salamander (Desmognathus marmoratus) in the southern Appalachians.
Molecular phylogenetics and evolution 38, 280-287.
Kozak KH, Blaine RA, Larson A (2006) Gene lineages and eastern North American
palaeodrainage basins: phylogeography and speciation in salamanders of the
Eurycea bislineata species complex. Molecular Ecology 15, 191-207.
Larson A, Wake DB, Yanev KP (1984) Measuring gene flow among populations
having high levels of genetic fragmentation. Genetics 106, 293-308.
Lefébure T, Douady CJ, Gouy M, et al. (2006) Phylogeography of a subterranean
amphipod reveals cryptic diversity and dynamic evolution in extreme
environments. Molecular Ecology 15, 1797-1806.
Levin HL (2009) The earth through time John Wiley & Sons.
Leys R, Watts CHS, Cooper SJB, Humphreys WF (2003) Evolution of subterranean
diving beetles (Coleoptera: Dytiscidae hydroporini, bidessini) in the arid zone
of Australia. Evolution 57, 2819-2834.
Liu L, Pearl DK (2007) Species Trees from Gene Trees: Reconstructing Bayesian
Posterior Distributions of a Species Phylogeny Using Estimated Gene Tree
Distributions. Systematic Biology 56, 504-514. 80
Lucas L, Gompert Z, Ott J, Nice C (2009) Geographic and genetic isolation in spring-
associated Eurycea salamanders endemic to the Edwards Plateau region of
Texas. Conservation Genetics 10, 1309-1319.
Meirmans PG, Van Tienderen PH (2004) GENOTYPE and GENODIVE: two programs
for the analysis of genetic diversity of asexual organisms. Molecular Ecology
Notes 4, 792-794.
Mueller RL (2006) Evolutionary rates, divergence dates, and the performance of
mitochondrial genes in Bayesian phylogenetic analysis. Systematic Biology
55, 289-300.
Niemiller ML, Fitzpatrick BM, Miller BT (2008) Recent divergence with gene flow in
Tennessee cave salamanders (Plethodontidae: Gyrinophilus) inferred from
gene genealogies. Molecular Ecology 17, 2258-2275.
Petranka J (1998) Salamanders of the United States and Canada Smithsonian
Institution Press, Washington.
Phillips CA (1994) Geographic distribution of mitochondrial DNA variants and the
historical biogeography of the Spotted Salamander, Ambystoma maculatum
Evolution 48, 597-607.
Porter M (2007) Subterranean biogeography: What have we learned from molecular
techniques? Journal of Cave and Karst Studies 69, 179-186.
Posada D (2008) jModelTest: phylogenetic model averaging. Molecular Biology and
Evolution 25, 1253-1256.
Pritchard JK, Stephens M, Donnelly P (2000) Inference of population structure using
multilocus genotype data. Genetics 155, 945-959. 81
Purrenhage JL, Niewiarowski PH, Moore FB (2009) Population structure of spotted
salamanders (Ambystoma maculatum) in a fragmented landscape. Mol Ecol
18, 235-247.
Raymond M, Rousset F (1995) GENEPOP (version 1.2): population genetics software
for exact tests and ecumenicism. Journal of Heredity 86, 248-249.
Rissler L (2010) Mapping amphibian contact zones and phylogeographical break
hotspots across the United States. Molecular Ecology 19, 5404-5416.
Rogers AR (1995) Genetic Evidence for a Pleistocene Population Explosion.
Evolution 49, 608-615.
Ronquist F, Teslenko M, Van Der Mark P, et al. (2012) MrBayes 3.2: Efficient
bayesian phylogenetic inference and model choice across a large model
spece. Systematic Biology.
Rosenberg NA (2003) distruct: a program for the graphical display of population
structure. Molecular Ecology Notes 4, 137-138.
Rousset F (2008) Genepop'007: a complete re-implementation of the genepop
software for Windows and Linux. Molecular ecology resources 8, 103-106.
Sambrook J, Fritsch E, T M (1989) Molecular Cloning: A laboratory Manual Cold
Springs Harbor Laboratory Press, New York.
Sket B (2008) Can we agree on an ecological classification of subterranean animals?
Journal of Natural History 42, 1549-1563.
Soltis DE, Morris AB, McLachlan JS, Manos PS, Soltis PS (2006) Comparative
phylogeography of unglaciated eastern North America. Molecular Ecology 15,
4261-4293. 82
Stamatakis A (2006) RAxML-VI-HPC: maximum likelihood-based phylogenetic
analyses with thousands of taxa and mixed models. Bioinformatics 22, 2688-
2690.
Stearn CW, Carroll RL, Clark TH (1979) Geological Evolution of North America John
Wiley & Sons.
Strecker U, Faúndez VH, Wilkens H (2004) Phylogeography of surface and cave
Astyanax (Teleostei) from Central and North America based on cytochrome b
sequence data. Molecular phylogenetics and evolution 33, 469-481.
Taylor DR, Keller SR, Funk D (2007) Historical range expansion determines the
phylogenetic diversity introduced during contemporary species invasion.
Evolution 61, 334-345.
Webb TI, Bartlein PJ (1992) Global Changes During the Last 3 Million Years: Climatic
Controls and Biotic Responses. Annual Review of Ecology and Systematics 23,
141-173.
Wessel A, Erbe P, Hoch H (2007) Pattern and process: Evolution of troglomorphy in
the cave-planthoppers of Australia and Hawai'i–Preliminary observations
(Insecta: Hemiptera: Fulgoromorpha: Cixiidae). Acta Carsologica 36, 199-
206.
Xia X, Lemey P (2009) Assessing substitution saturation with DAMBE. In: The
Phylogenetic Handbook: A Practical Approach to DNA and Protein Phylogeny
(eds. Lemey P, Salemi M, Vandamme A). Cambridge University Press,
Cambridge, United Kingdom. 83
Xia X, Xie Z, Salemi M, Chen L, Wang Y (2003) An index of substitution saturation
and its application. Molecular phylogenetics and evolution 26, 1-7.
Zakšek V, Sket B, Trontelj P (2007) Phylogeny of the cave shrimp Troglocaris:
Evidence of a young connection between Balkans and Caucasus. Molecular
phylogenetics and evolution 42, 223-235.
Zamudio KR, Savage WK (2003) Historical isolation, range expansion, and
secondary contact of two highly divergent mitochondrial lineages in Spotted
Salamanders (Ambystoma maculatum). Evolution 57, 1631-1652.
84
Figures
Figure 1. Depiction of the range of Eurycea lucifuga (shown in grey), and collection locations (black circles). Size of the location marker indicates number of samples collected, which ranged from 1-11.
1" 2"
W" C" E" W" E" C"
3" 4"
C" E" W" C" W" E"
Figure 2. The four scenarios tested in the DIYABC model, depicting hypothetical relationships among the
Eastern, central, and Western populations. On the right the results of a logistical model comparing the posterior probability of each scenario with the number of simulations used to calculate it. 85
E_longicauda Su7_20_1 E7_20_2 G6_5_1 E7_20_1 Pr6_6_1 Oe6_7_2 Cr6_7_4 Oe6_7_4 Oe6_7_1 Oe6_7_8 E7_20_4 D7_19_3 G6_5_2 Pr6_6_6 D7_19_1 Su7_20_5 D7_19_5 E7_20_6 RC7_19_2 RC7_19_4 Central% Su7_20_3 E7_20_5 (KY%and% Pr6_6_5 IN)% Su7_20_2 BF7_18_9 E7_20_3 BF7_18_6 BF7_18_8 RL7_18_1 RL7_18_4 BF7_18_1 Cr6_7_2 RC7_19_1 BF7_18_7 RC7_19_3 RL7_18_3 BF7_18_2 BF7_18_3 D7_19_2 D7_19_4 Pom6_6_2 Pom6_6_3 Smoke2 Taw9_24_3 BLB9_4_6 Byrd10_1_16 BLB9_4_1 Smoke9_3_4 Byrd10_1_3 Man7_28_1 East% Man7_28_2 Byrd10_1_4 (VA,% H6_13_1 Spr7_28_1 WV%and% Byrd10_1_6 TN)% Taw9_24_1 Byrd10_1_11 Smoke1 BLB9_4_2 LP6_3_4 Man7_28_3 Bor7_28_1 Jan7_9_1 J7_10_5 IG7_10_3 J7_10_4 Cs7_13_6 Bl7_11_1 Virginia' Bu7_14_5 Th7_11_3 W.'Virginia' Th7_11_4 Jan7_9_2 Jan7_9_5 West% E.'Tennessee' Jan7_9_7 Bu7_14_7 (OK%and% Indiana' Cs7_13_5 Th7_11_2 MO)% Tennessee' Bu7_14_3 Cs7_13_1 Oklahoma' Bu7_14_6 Cs7_13_4 Missouri' J7_10_1 Bu7_14_1 Bu7_14_9 Cs7_13_2 Cs7_13_7
0.4 Figure 3. Bayesian phylogeny inferred from the concatenated sequence data of cytb, ND2 and POMC using
MrBayes. Pseudotriton ruber was included as an outgroup, but is not shown here. Posterior support values above 0.5 are shown above, and maximum likelihood bootstraps above 50 are shown below branches.
Colors correspond to the regions from which samples were collected, as indicated in the legend.
86
E.#longicauda#
1 0.353 Western'
1 4.318 3.804
1 2.694
0.745 0.893 Central'
0.684 1.687
0.479 0.658 Eastern'
Time'since'present'(millions'of'years)'
5.0 4.0 3.0 2.0 1.0 0.0 Figure 4. Dates of divergence, inferred using the species tree method in BEAST with a strict clock,
indicating divergence of the major lineages during the late Pliocene and early Pleistocene. Node labels
reflect the estimates ages of divergence, and branch labels represent the posterior probabilities.
87
Mann
Borehole
Mann
Spring
Culver Buckeye_Creek
Higganbotham Borehole
Spring Culver
Smoke
Buckeye_Creek
Tawney Higganbotham Smoke
Blankenship
Tawney
Byrd Eastern)(WV/VA)) Blankenship
Byrd Puddle
Bankley Puddle
Bankley Robinson
Roberts
Robinson
Sullivan Roberts
Donnehue
Central)(IN)) Sullivan
Lost
Donnehue
Lost
YMCA
YMCA White
Violet White
Sturgeon
Violet Stan
Stan Silver
Sturgeon
Silver Phil
Phil Pagoda
Pagoda
New
New
Natural
Little
Left
Historical
Natural Little Left
Hickory
Historical
Hickory
Great_Onyx
Great_Onyx Fall
Fall
Crystal Central)(KY/TN))
Cadaverous
Crystal
C2
Cadaverous C2
Black
Big Black
Big
Austen
Adwell
Crews
Austen
Adwell
Eoff Crews
Eoff
Gillespie
Prowell Gillespie
Pompie
Prowell
Pompie Bull_Creek
Crighton Bull_Creek
Third
Crighton
Survivalist Third
Blue_Moon
January Survivalist
Blue_Moon
January
Jail Western)(OK/MO)) Jail
Iron_Gate Iron_Gate Individual)) Q)scores) Popula2on) Q)mean) scores)
Figure 5. The geographic clustering of diversity, estimated using STRUCTURE, indicates four major clades across the range of Eurycea lucifuga. While the Eastern and Western clades are distinct from each other entirely, they both contribute to the diversity found within the two central clades, indicating either asymmetric gene flow or past expansion from the central region. 88
Tables
Table 1. Results of demographic analysis using Tajima’s D and Fu’s F statistics, as well as Mismatch analysis, indicate that recent expansion is likely within each region, on a time scale that supports divergence dates found in BEAST species tree inference.
Major divergence: Western Central Eastern and Western Tajima’s D -1.85* -0.67 -2.07* Fu’s F -19.70* -24.07* -1.43
Tau 23.37 / 5.02 0.24 / 0.13 6.35 / 10.54
Theta0 0 / 0 13.89/ 0 0 / 0
Theta1 3.17 / 2.42 99,999.00 / 99,999.00 6.88 / 12.69
SSD 0.015 0.006 0.044 R 0.017 0.003 0.059 Evidence for Recent Expansion? Yes Yes Yes Estimated time since expansion: 3.3my/1.2my 34,000y/33,000y 909,600y/2.58my
Table 2. A comparison of the posterior probabilities of each of four divergence scenarios analyzed in
DIYABC calculated using 4000 simulations, presented with confidence intervals. The highest probability, that of scenario three, is in bold.
Posterior Probability 95% CI Scenario 1 0.3055 0.2241,0.3869 Scenario 2 0.0762 0.0055,0.1469 Scenario 3 0.5154 0.4494,0.5815 Scenario 4 0.1029 0.0313,0.1745
89
Table 3. Comparisons of heterozygosity, inbreeding values, and distance statistics among the four main lineages in Eurycea lucifuga. Heterozygosity is significantly greater in the central regions than the East or
West, supporting a hypothesis of expansion from the central region.
Statistic Western Central-KY and TN Central-IN Eastern OSx P-value Ho 0.165 0.217 0.183 0.114 0.149 0.003 Hs 0.163 0.269 0.226 0.163 0.18 0.003 Gis -0.014 0.195 0.192 0.303 0.459 0.012 Gst 0.063 0.062 0.016 -0.089 0.247 0.641 D_est 0.015 0.025 0.006 -0.018 0.063 0.929
90
Table 4. AMOVAs across the range as well as within each region indicate that the majority of genetic variation is partitioned among regions, and that regions differ widely in the amount of population structure they exhibit.
Range-wide
Source Nested in %var F-stat F-value Std.Dev. c.i.2.5% c.i.97.5% P-value Within Individual -- 0.613 R_it 0.387 0.26 0.164 0.705 -- Among Individual Population 0.014 R_is 0.022 0.233 -0.111 0.435 0.279 Among Population Series_A 0.038 R_sc 0.057 0.03 0.009 0.074 0.001 Among Region -- 0.335 R_ct 0.335 0.188 0.16 0.593 0.001 Western
Source Nested in %var F-stat F-value Std.Dev. c.i.2.5% c.i.97.5% P-value Within Individual -- 1.184 R_it -0.184 0.487 -0.333 0.609 -- Among Individual Population -0.201 R_is -0.204 0.511 -0.368 0.619 0.942 Among Population -- 0.017 R_st 0.017 0.023 -0.043 0.06 0.249 South-Central
Source Nested in %var F-stat F-value Std.Dev. c.i.2.5% c.i.97.5% P-value Within Individual -- 0.829 R_it 0.171 0.094 0.056 0.432 -- Among Individual Population 0.087 R_is 0.095 0.144 -0.023 0.427 0.039 Among Population -- 0.084 R_st 0.084 0.06 0.004 0.127 0.001 Indiana
Source Nested in %var F-stat F-value Std.Dev. c.i.2.5% c.i.97.5% P-value Within Individual -- 1.037 R_it -0.037 0.311 -0.216 0.454 -- Among Individual Population -0.092 R_is -0.098 0.337 -0.292 0.421 0.842 Among Population -- 0.055 R_st 0.055 0.018 -0.001 0.1 0.046 Eastern
Source Nested in %var F-stat F-value Std.Dev. c.i.2.5% c.i.97.5% P-value Within Individual -- 0.859 R_it 0.141 0.289 0.02 0.679 -- Among Individual Population 0.147 R_is 0.146 0.264 0.044 0.681 0.123 Among Population -- -0.006 R_st -0.006 0.042 -0.036 0.057 0.534
91
Supplementary Figures
E_longicauda BF2 60# Ad7_1_1 1# D2 BF3 RL3 D4 Pom2 Pom3 RC3 Su2 E3 BF6 RC4 Pr5 E6 D5 BF8 RL7_18_2 Stan7_3_1 RL1 63# Cr2 RC1 BF1 RC2 Central# Pag7_6_3 Pag7_6_2 (IN#and# Bla7_3_1 D1 TN)# Su5 Oe8 D3 G2 Pr6 Smoke2 E4 Lil7_2_1 Su1 E2 G1 E1 Pr6_6_2 Pr1 Taw3 Cr4 G6_5_4 Oe1 0.68# Cr6_7_3 Oe6_7_7 Oe2 Spr1 BLB1 H2 Byrd10_1_13 Taw6_5_1 Eastern# Byrd10_1_14 Smoke1 (VA#and# 1# BC7_28_1 LP4 WV)# Byrd11 Man1 HM7_28_1 Cu6_13_2 Byrd3 Bor6_13_1 90# BLB6 1# Byrd16 53#IG3 57#J4 IG7_10_1 1# J7_10_2 IG7_10_2 Jan7_9_3 Cs2 Bu6 79# Bu9 60# Bu1 Western# 0.97#Cs4 J1 (OK#and# Th3 64# Bl1 MO)# S7_11_2 S7_11_1 Bl7_11_5 Cs1 Th7_11_1 Jan1 Bu7_14_2
0.07 Fig. S1. Maximum likelihood tree of cytb, on which the major geographic regions are labeled.
Pseudotriton ruber was included as an outgroup, but is not shown here. Branches are labeled with bootstraps above 50. Bayesian reconstructions supported major regional relationships, and posterior support values above 0.5 from the Bayesian reconstruction are included below branches where appropriate. 92
E_longicauda E_longicauda 93# IG3 IG3 1# J4 J4 Jan1 Jan1 78# Bl1 Bl1 Bu5 Bu5 Jan5 Western# Jan5 73# Th4 Th4 0.97# (OK#and# Cs1 Cs1 Bu3 MO)# Bu3 98# Jan2 Jan2 1# J1 J1 Cs2 Cs2 Th2 Th2 Bu1 Bu1 Bor6_13_1 Bor6_13_1 Byrd4 Byrd4 H6_13_2 H6_13_2 55# Byrd3 Byrd3 HM7_28_1 HM7_28_1 59# Smoke1 Smoke1 98# Spr1 Eastern# Spr1 1# Bor6_13_2 Bor6_13_2 Man1 (VA#and# Man1 LP4 WV)# LP4 Cu6_13_2 Cu6_13_2 BLB1 BLB1 BC7_28_1 BC7_28_1 Taw1 Taw1 H2 H2 G6_5_3 G6_5_3 80# Oe4 Oe4 1# Pr1 Pr1 Cr2 Cr2 Pom2 Pom2 G2 G2 Cr6_7_3 Cr6_7_3 Oe2 Oe2 G1 G1 Pom3 Pom3 Pr6_6_3 Pr6_6_3 Pr6_6_4 Pr6_6_4 Oe8 Oe8 Oe1 Oe1 Cr6_7_5 Central# Cr6_7_5 Cr4 (IN#and# Cr4 E5 E5 Su1 TN)# Su1 RL1 RL1 RC1 RC1 E1 E1 D2 D2 BF1 BF1 Pr5 Pr5 D1 D1 Su7_20_6 Su7_20_6 D3 D3 BF7_18_4 BF7_18_4 BF2 BF2 RL4 RL4
0.2 0.2 Fig. S2. Maximum likelihood tree of Nd2, on which the major geographic regions are labeled.
Pseudotriton ruber was included as an outgroup, but is not shown here. Branches are labeled with
bootstraps above 50. Bayesian reconstructions supported major regional relationships, and posterior
support values above 0.5 from the Bayesian reconstruction are included below branches where appropriate. 93
E_longicauda E4 99" Su1 Su7_20_4 Su3 E5 D4 Pr6_6_4 Pr6_6_2 RC1 Smoke4 His6_29_1 D3 G2 Su2 BLB6 Byrd3 Oe6_7_7 BLB1 Byrd4 Cr6_7_5 Pom3 BF1 BF9 RC2 RL3 E1 Aus7_2_1 Cr2 RL1 Oe4 Taw1 Man1 E2 Man2 Oe2 G1 BLB9_4_7 Oe8 Spr1 Pr6_6_3 G6_5_3 89" Smoke9_3_1 J7_10_8 HM7_28_1 RL4 Bl7_11_5 BLB2 Oe1 D1 Left7_3_1 H2 BF6 Bor1 Stur7_3_1 Crys6_30_2 Lil7_2_1 Su7_20_6 Man3 Taw6_5_1 S7_11_2 Bu6 E6 Pom2 Smoke1 Pr1 His6_29_2 Stan7_3_1 RC4 Byrd6 Byrd11 Ad7_1_1 Jan7 65" IG3 Phil7_3_4 J1 D2 BF7 Bla7_3_1 Bl1 Th2 Jan1 J7_10_9 Aus6_30_1 Cs1 LP4 H6_13_2 Bu1
0.0050 Fig. S3. Maximum likelihood tree of POMC. Pseudotriton ruber was included as an outgroup, but is not shown here. Major regional groups were recovered from POMC data in neither the Maximum Likelihood nor the Bayesian tree. Branches are labeled with bootstraps above 50. 94
Supplementary Tables
Table S1. Primer sources and sequence for both PCR and sequencing steps, as well as thermocycler conditions for each primer set. Primer Source Sequence Cycling conditions cytbF Harlan and Zigler AAGATTATTAATAACTCCTTTATTGA Annealing temp. of 50C, (2009) 35 cycles cytbR AAAATGCTTGTCCAATTTCAAT
ND2F This study TACAAGCCTCAGCATCTGCC Annealing temp. of 59.4C, 30 cycles ND2R ATCCAGAGGTTGGTGGGAGT
POMCF Lamb et al. ATATGTCATGAGCCATTTTCGCTGGAA Annealing temp. of 58C, (2012) 45 cycles POMCR GGCATTTTTGAAAAGAGTCATTAGAGG
Table S2.: Collection efforts/results Sample ID Species State Cave Location Date Ad7.1.1 Eurycea lucifuga Kentucky Adwell 7/1/13 BF7.18.1 Eurycea lucifuga Indiana Bankley/Fairground 7/18/12 BF7.18.2 Eurycea lucifuga Indiana Bankley/Fairground 7/18/12 BF7.18.3 Eurycea lucifuga Indiana Bankley/Fairground 7/18/12 BF7.18.6 Eurycea lucifuga Indiana Bankley/Fairground 7/18/12 BF7.18.7 Eurycea lucifuga Indiana Bankley/Fairground 7/18/12 BF7.18.8 Eurycea lucifuga Indiana Bankley/Fairground 7/18/12 BF7.18.9 Eurycea lucifuga Indiana Bankley/Fairground 7/18/12 Bl7.11.1 Eurycea lucifuga Oklahoma Blue Moon 7/11/12 Bla7.3.1 Eurycea lucifuga Kentucky Black Rock 7/3/13 BLB9.4.1 Eurycea lucifuga Virginia Blankenship Blowhole 9/4/11 BLB9.4.2 Eurycea lucifuga Virginia Blankenship Blowhole 9/4/11 BLB9.4.6 Eurycea lucifuga Virginia Blankenship Blowhole 9/4/11 Bor7.28.1 Eurycea lucifuga West Virginia Boarhole 7/29/12 Bu7.14.1 Eurycea lucifuga Missouri Bull Creek 7/14/12 Bu7.14.3 Eurycea lucifuga Missouri Bull Creek 7/14/12 Bu7.14.5 Eurycea lucifuga Missouri Bull Creek 7/14/12 Bu7.14.6 Eurycea lucifuga Missouri Bull Creek 7/14/12 Bu7.14.7 Eurycea lucifuga Missouri Bull Creek 7/14/12 Bu7.14.9 Eurycea lucifuga Missouri Bull Creek 7/14/12 Byrd10.1.11 Eurycea lucifuga Virginia Byrd's water 10/1/11 Byrd10.1.16 Eurycea lucifuga Virginia Byrd's water 10/1/11 Byrd10.1.3 Eurycea lucifuga Virginia Byrd's water 10/1/11 Byrd10.1.6 Eurycea lucifuga Virginia Byrd's water 10/1/11 Cr6.7.2 Eurycea lucifuga Tennessee Crews 6/7/12 Cr6.7.4 Eurycea lucifuga Tennessee Crews 6/7/12 95
Cr7.13.1 Eurycea lucifuga Missouri Crighton Spring 7/13/12 Cr7.13.2 Eurycea lucifuga Missouri Crighton Spring 7/13/12 Cr7.13.4 Eurycea lucifuga Missouri Crighton Spring 7/13/12 Cr7.13.5 Eurycea lucifuga Missouri Crighton Spring 7/13/12 D7.19.1 Eurycea lucifuga Indiana Donnehue 7/19/12 D7.19.2 Eurycea lucifuga Indiana Donnehue 7/19/12 D7.19.3 Eurycea lucifuga Indiana Donnehue 7/19/12 D7.19.4 Eurycea lucifuga Indiana Donnehue 7/19/12 D7.19.5 Eurycea lucifuga Indiana Donnehue 7/19/12 E7.20.1 Eurycea lucifuga Indiana Lost Cave 7/20/12 E7.20.2 Eurycea lucifuga Indiana Lost Cave 7/20/12 E7.20.3 Eurycea lucifuga Indiana Lost Cave 7/20/12 E7.20.4 Eurycea lucifuga Indiana Lost Cave 7/20/12 E7.20.5 Eurycea lucifuga Indiana Lost Cave 7/20/12 E7.20.6 Eurycea lucifuga Indiana Lost Cave 7/20/12 G6.5.1 Eurycea lucifuga Tennessee Gillespie 6/5/12 G6.5.2 Eurycea lucifuga Tennessee Gillespie 6/5/12 H6.13.1 Eurycea lucifuga West Virginia Higganbotham 6/13/13 IG7.10.3 Eurycea lucifuga Oklahoma Iron Gate 7/10/12 J7.10.1 Eurycea lucifuga Oklahoma Jail 7/10/12 J7.10.4 Eurycea lucifuga Oklahoma Jail 7/10/12 J7.10.5 Eurycea lucifuga Oklahoma Jail 7/10/12 Jan7.9.1 Eurycea lucifuga Oklahoma January-Stansbury 7/9/12 Jan7.9.2 Eurycea lucifuga Oklahoma January-Stansbury 7/9/12 Jan7.9.5 Eurycea lucifuga Oklahoma January-Stansbury 7/9/12 Jan7.9.7 Eurycea lucifuga Oklahoma January-Stansbury 7/9/12 LP6.3.4 Eurycea lucifuga Tennessee Lost Puddle 6/3/12 Man7.28.1 Eurycea lucifuga West Virginia Mann 7/29/12 Man7.28.2 Eurycea lucifuga West Virginia Mann 7/29/12 Man7.28.3 Eurycea lucifuga West Virginia Mann 7/29/12 Oe6.7.1 Eurycea lucifuga Tennessee Eoff 6/7/12 Oe6.7.2 Eurycea lucifuga Tennessee Eoff 6/7/12 Oe6.7.4 Eurycea lucifuga Tennessee Eoff 6/7/12 Oe6.7.8 Eurycea lucifuga Tennessee Eoff 6/7/12 Pom6.6.2 Eurycea lucifuga Tennessee Pompie 6/6/12 Pom6.6.3 Eurycea lucifuga Tennessee Pompie 6/6/12 Pr6.6.1 Eurycea lucifuga Tennessee Prowell 6/6/12 Pr6.6.5 Eurycea lucifuga Tennessee Prowell 6/6/12 Pr6.6.6 Eurycea lucifuga Tennessee Prowell 6/6/12 RC7.19.1 Eurycea lucifuga Indiana Roberts 7/19/12 RC7.19.2 Eurycea lucifuga Indiana Roberts 7/19/12 RC7.19.3 Eurycea lucifuga Indiana Roberts 7/19/12 RC7.19.4 Eurycea lucifuga Indiana Roberts 7/19/12 96
RL7.18.1 Eurycea lucifuga Indiana Robinson Ladder 7/18/12 RL7.18.2 Eurycea lucifuga Indiana Robinson Ladder 7/18/12 RL7.18.3 Eurycea lucifuga Indiana Robinson Ladder 7/18/12 S7.11.1 Eurycea lucifuga Oklahoma Survivalist 7/11/12 S7.11.2 Eurycea lucifuga Oklahoma Survivalist 7/11/12 Smoke1 Eurycea lucifuga Virginia Smokehole 9/3/11 Smoke2 Eurycea lucifuga Virginia Smokehole 9/3/11 Smoke9.3.4 Eurycea lucifuga Virginia Smokehole 9/3/11 Spr7.28.1 Eurycea lucifuga West Virginia Spring 7/29/12 Stan7.2.1 Eurycea lucifuga Kentucky Stan's Well 7/2/13 Su7.20.1 Eurycea lucifuga Indiana Sullivan 7/20/12 Su7.20.2 Eurycea lucifuga Indiana Sullivan 7/20/12 Su7.20.3 Eurycea lucifuga Indiana Sullivan 7/20/12 Su7.20.4 Eurycea lucifuga Indiana Sullivan 7/20/12 Su7.20.5 Eurycea lucifuga Indiana Sullivan 7/20/12 Su7.20.6 Eurycea lucifuga Indiana Sullivan 7/20/12 Taw9.24.1 Eurycea lucifuga Virginia Tawney's 9/24/11 Taw9.24.2 Eurycea lucifuga Virginia Tawney's 9/24/11 Taw9.24.3 Eurycea lucifuga Virginia Tawney's 9/24/11 Th7.11.1 Eurycea lucifuga Oklahoma Third 7/11/12 Th7.11.2 Eurycea lucifuga Oklahoma Third 7/11/12 Th7.11.3 Eurycea lucifuga Oklahoma Third 7/11/12 Th7.11.4 Eurycea lucifuga Oklahoma Third 7/11/12
97
CHAPTER 3
POTENTIAL SELECTIVE MAINTENANCE OF COLOR THROUGH NON-CAVE DISPERSAL IN THE CAVE SALAMANDER, EURYCEA LUCIFUGA (PLETHODONTIDAE)
Abstract
Coloration is important in vertebrate physiological function and communication.
The evolution of color in amphibians is thought to be driven largely by predator pressure, which has resulted in bright red and orange patterning in aposematic species and their mimics. In lineages that colonized caves coloring is often lost as a result of relaxed selection, presumably due to lack of light. However, some cave-dwelling salamanders that are not restricted to the cave environment have retained their coloration. We tested for evidence that natural selection maintains the coloration of a brightly colored cave- dwelling salamander, Eurycea lucifuga, on different geographic scales: across its range and within a single large cave system. We found through a comparison of phenotypic and neutral genetic differentiation that color variation exhibits different spatial structuring than neutral loci, suggesting that it has been influenced by non-neutral processes.
Furthermore, genetic analyses suggest that dispersal among caves occurs predominantly over the surface habitat, which may contribute to selective maintenance of visual traits such as coloration. These results indicate that occasional exposure to environmental selection pressures may influence trait evolution, and also emphasize the importance of choosing appropriate or multiple spatial scales at which to examine evolutionary and ecological dynamics. 98
Introduction
Color patterning has been extensively studied because of its variability and also its clear functional importance to many taxa. Amphibian systems have been the focus of many such studies due to the diversity of coloration even among closely related taxa, and the dynamic selection regimes that characterize coloration evolution (Rudh &
Qvarnström 2013). Among amphibian systems, main functions of coloration include thermoregulation, UV protection, communication to mates, and avoidance of predation through crypsis, active defense behaviors, or aposematism (Rudh & Qvarnström 2013).
The last of these, aposematism, is thought to be the driver behind some of the most dramatic displays of coloration, and has been the inspiration of research into the proximate and ultimate mechanisms of this type of predator avoidance. Aposematic coloration advertises an animal’s toxicity to predators conspicuously, including colors, odors, and behavior (Lindström 1999; Poulton 1890).
Warning coloration is an important strategy for amphibians, since they have small bodies and lack protective structures, and therefore are vulnerable to predation (Rudh &
Qvarnström 2013). Although coloration displays are variable, typical warning colors are red, orange, yellow, and white (Lindström 1999). These colors are produced by three types of specialized pigment-containing epidermal cells called chromatophores: xanthophores, which produce yellow and red pigmentation, iridophores, which contain reflecting platelets, and melanophores, which are responsible for darker pigmentations
(Rudh & Qvarnström 2013).
Although salamanders tend to be largely nocturnal, it is not uncommon to observe diurnal activity in most species (Brandon & Huheey 1975), and they represent valuable 99 prey for a wide variety of predators (Petranka 1998). However, the combination of aposematism and a tendency to hide in dry weather make salamanders infrequent prey for some birds, such as the Hermit thrush and barred and screech owls (Jaeger 1981).
Nonetheless, the risk of predation is common enough for both aposematism and mimicry of aposematic species to have developed. A common aposematism/mimicry paradigm is that of the ‘red salamanders’ found in many regions of North America. For example, the brightly colored species Notophthalmus viridescens, which is highly toxic, lives in sympatry with two Mullerian mimics which benefit from its warning coloration,
Pseudotriton ruber (Brandon et al. 1979; Brodie 1981; Howard & Brodie 1971) and the red erythristic phase of Plethodon cinereus (Tilley et al. 1982). Similarly, Taricha newts in the western United States have influenced the mimetic coloration of Ensatina eschscholtzii xanthoptica (Kuchta & Reeder 2005). Bird predators are able to discriminate between red aposematic or mimic species and their cryptic counterparts
(Kraemer & Adams 2014), and red individuals still experience significantly less predation than other morphs (Brodie Jr & Brodie III 1980; Tilley et al. 1982). Mimicry in salamanders also requires less precision than has been found in butterfly or wasp-fly systems (Howard & Brodie 1973).
The variety of color patterns exhibited by many salamander species makes the lack of pigmentation in cave-restricted species stand out in stark contrast. There are approximately twelve species and subspecies of cave-obligate salamanders in the contiguous United States, most within the Plethodontid genus Eurycea (Culver et al.
2000). Cave obligates are well known for the parallel evolution of a suite of traits known as troglomorphy, which include both regressive features such as depigmentation and eye 100 reduction and constructive features such as enhanced extra-optic sensory structures
(Porter 2007). The evolutionary mechanisms responsible for these traits have been well characterized in Astyanax cave fish. While some troglomorphic traits such as eye reduction and sensory structures arise through directional selection (Jeffery et al. 2003;
Jeffery 2009; Yamamoto et al. 2004), depigmentation is apparently the result of relaxed selection and subsequent accumulation of loss of function mutations (Protas et al. 2007).
Despite the repeated evolution of troglomorphy in cave-dwelling taxa (Jones et al. 1992), there remains a class of cave species that has not acquired troglomorphic traits. These species, known as troglophiles, maintain permanent populations in caves but anecdotal reports suggest they can visit niches outside of caves in the surface environment (Sket
2008).
Hypotheses regarding why some species become morphologically troglomorphic while others do not include the amount of time a species has inhabited the cave environment, a tendency of troglophiles to live closer to cave entrances where presumably they would be exposed to light conditions, and developmental constraints on troglomorphic features (Pipan & Culver 2012). It has also been suggested that inbreeding depression prevents the formation of viable troglomorphic population because of their isolation and small population numbers (Avise & Selander 1972). Some of these hypotheses are not well supported: for example, the prevalence of young troglomorphic populations (e.g. Astyanax fish (Strecker et al. 2004), Cixiidae plant hoppers (Wessel et al. 2007), Nesticella spiders (Zhang & Li 2013)) leads to the conclusion that a cave lineage’s age is not correlated with its degree of troglomorphy (Wessel et al. 2007).
Exposure to a photic environment such as in the entrances of caves does seem like a 101 likely contributing factor to an avoidance of troglomorphy, though troglophilic salamanders show a preference for deeper, darker, more humid caves (Ficetola et al.
2012, 2013; Lunghi et al. 2014). Exposure to light may be more common in troglophiles due to their tendency to move among caves more than troglomorphic cave species
(Caccone 1985). Though troglomorphic species do occasionally leave the caves, particularly at night (Schlegel et al. 2009), common use of the surface habitat during dispersal by troglophiles may impose selection pressure on coloration as a predator defense strategy, leading to its persistence in troglophilic populations.
In this study, we explore the maintenance of pigmentation in troglophiles. We tested whether selection influences coloration by comparing phenotypic differentiation and neutral genetic variation in a troglophilic salamander, Eurycea lucifuga at both broad and fine spatial scales. In order to determine whether exposure to the surface environment plays a part in the selective maintenance of color, we also characterized dispersal within a large cave system, focusing on whether gene flow occurs across the surface habitat as opposed to through subterranean routes.
Methods
Study system
Eurycea lucifuga, the Cave Salamander, is a troglophilic salamander commonly found in caves throughout the Central and Southeast United States. It belongs to a genus in which cave-dwelling and troglomorphy have evolved independently multiple times
(Bonett et al. 2014). The coloration of E. lucifuga as adults is very similar to that of other ‘red salamanders’: the body is bright orange, with black spots covering the entire surface (Fig. 1). It is unknown whether, like most members of Plethodontinae (Brodie 102
1977), E. lucifuga is noxious to predators; however, the larvae are palatable to certain fish predators (Kats et al. 1988). In addition to color patterning, E. lucifuga has a characteristic anti-predator posture similar to that of many Plethodontids, in which the animals contort their torso into a circular posture while pointing their tail straight up, and undulating the tail. This behavior is effective in deterring predation by shrews by directing them to the tail, which when bitten exposes the predator to the noxious secretions and is autotomizable (Brodie et al. 1979). Though E. lucifuga spends the majority of its adult life in the crevices of cave floors and walls, and is obligately associated with caves during the egg and larval stages, anecdotal reports suggest it is occasionally found outside of caves (Hutchison 1958; Petranka 1998).
Sample collection and data compilation
Over the months of May through July during 2012-2014 we sampled 1-12 individuals of Eurycea lucifuga from 27 populations located in Tennessee, Kentucky,
Oklahoma, Missouri, and Indiana for a total of 153 individuals (Fig. 2; Table S1). We first collected tail clips to genotype each individual. We also photographed each individual with a Pentax Optio WG-2 digital camera from a distance of twelve inches with the flash on. In each photograph we included as a color standard a paint sample card
(color BHG511).
Tail tissues were stored at room temperature in 95% ethanol until total genomic
DNA was extracted from each using either a phenol-chloroform (Sambrook et al. 1989) or chelex (using a 5-10% slurry of Chelex and an incubation time of 180 minutes at
95°C) extraction method (Walsh et al. 1991). We genotyped each individual at 19 microsatellite loci developed for this species (Appendix A) by amplification in 103 multiplexed reactions using the Polymerase Chain Reaction (PCR), and products were sent to the DNA Analysis Facility on Science Hill at Yale University (New Haven, CT) for fragment analysis using a 3730xl 96-capillary Genetic Analyzer with the DS-33 dye set. We used the software GeneMarker (SoftGenetics LLC, State College, PA) to call alleles at each locus for each individual.
We digitally analyzed the photographs using the software program ImageJ
(Rasband 1997-2014) First we measured values of Red, Green and Blue (RGB) from an area in the orange background skin, from a black spot, and from the color standard in the photograph. Color measurements on each salamander were made on the dorsum between the fore- and hindlimbs; care was taken to avoid light fluctuations such as shine spots. All
RGB measurements were converted to Hue, Saturation, and Value (HSV) using the function rgb2hsv() in the R statistical package rgDevices (R Core Team), in R v3.1.2 (R
Development Core Team, 2008) interfaced with RStudio v.00.98.1091(RStudio, Inc.).
We regressed HSV for orange background and black spot onto the corresponding standard trait using the function lm() in the R statistical package stats (R Core Team) to obtain the residuals for further analysis. Following the example of Zhou et al. (2014) we perfromed Principal Components Analysis (PCA) on hue, saturation, and value for both the orange background and black spots in order to examine a trait for each representing the greatest amount of variation. We used the function prcomp() in the R package stats.
Next, a rectangular area was selected on the dorsum of each salamander midway between the forelimb and hind limb junctions, approximately as wide as the salamander. We measured the area of all of the spots within that rectangle as well as the total area of the selection, and divided spot area by the total selection area to produce a black spot:orange 104 background ratio trait. Thus, the three traits we examine in the subsequent analyses were the first principal component (PC1) of the orange background (“orange PC1”), PC1 of the black spots (“black PC1”), and the ratio of spot area to total area (“black:orange ratio”).
The majority of variance (86%) in orange PC1 is represented by strong negative loadings of saturation and value, with little variation seen in the hue of the orange background.
The first principal component representing black PC1 comprises 65% of the variance, and represents strong positive loadings of all color components, hue, saturation, and value (Table 1).
To complement our range-wide sampling scheme with fine-scale sampling, we extensively sampled E. lucifuga within one local area, the Mammoth Cave National Park
(Edmonson, KY). We collected from 13 cave entrances, with the goal of capturing individuals from caves connected to the main Mammoth Cave system as well as those isolated from the system (Fig. 3). We obtained information about the subterranean connectedness of the cave entrances from park officials, and recorded whether or not they are separated by the Green River, which may present a potential barrier to above-ground dispersal. As with the range-wide data, we performed PCA to obtain traits representing the majority of variation in both the orange background and black spots. The first principal component representing variation in the orange background trait (“orange
PC1”) accounts for 96% of variance, and is largely representative of strong positive loadings on saturation and value. PC1 for the black spot trait (“black PC1”) accounts for
59% of variation, and represents equally strong loadings for each color component (Table
1). 105
Analyses of genetic and phenotypic variance
To determine population genetic structure, we used the program GenoDive
(Meirmans & Van Tienderen 2004) to produce a table of pairwise FST values for each population in the dataset, replacing any negative values with zero. Additionally, we used the function earth.dist() in the R statistical package fossil (Vavrek 2011) to produce a matrix of pairwise geographic distances.
We followed the methods of Antoniazza et al. (2010), Manier et al. (2007), and
Berardi et al. (In review) to calculate values of PST as estimations of phenotypic differentiation analogous to FST. This method is an adaptation of the FST-QST comparison, in which differentiation in neutral genetic variation is compared with a representation of among-population variation in a standardized quantitative trait (Spitze
1993). However, because calculations of QST require estimates of additive genetic variance that are often unavailable in wild populations, the phenotypic equivalent, PST, is used in substitution (Sæther et al. 2007). For each color trait we used the results of a linear model to populate two matrices containing among- and within-population pairwise mean squares. We divided the difference between these matrices by the weighted number of populations, which resulted in a matrix of between-population variance (σ2b).
These σ2b values were used with the within-population variances (σ2w, approximated by the within-population mean squares) to produce pairwise estimates of phenotypic structure, PST, using the following equation:
g σ!b P = !" (gσ!b) + (2h! σ!w)
2 where g is approximated by among-population genetic variance and h is heritability. PST is commonly used to detect evidence of local adaptation or stabilizing selection on 106 phenotypic traits (Antoniazza et al. 2010; Berardi et al. In review; Sæther et al. 2007;
Spitze 1993).
As neither g nor h2 have been estimated for this species, we looked to the literature for guidance on estimates to use in our calculations. Although heritability of color traits has not been estimated in any close relatives, the prediction that heritability in color traits in anurans is predicted to be high (Hoffman & Blouin 2004) led us to follow the example of other FST-PST comparisons in conducting our analyses with a heritability of 0.5 (Merilä & Crnokrak 2001; Storz 2002; Wilson et al. 2013). As other studies comparing phenotypic and neutral genetic structure have used additive genetic variances of 1 (Alho et al. 2010; Antoniazza et al. 2010; Leinonen et al. 2006; Manier et al. 2007), and this is within the range observed in Hansen (2011). Thus, we used this estimation of g=1. We therefore proceeded with a heritability of 0.5 and additive genetic variance of 1 for our traits, but interpretation of these results must take into account that much higher or much lower heritability and additive genetic variance could affect the results of our analyses.
To understand the implications of these assumptions on our results, we performed sensitivity analyses of our PST calculations by permuting them using three values of g
(0.01, 0.1, and 1) and five values of h2 (0.1, 0.25, 0.5, 0.75, and 1) (Antoniazza et al.
2010; Berardi et al. In review; Manier et al. 2007). This gives us an understanding of how varying these parameters will impact our analyses.
Statistical Analyses: Gene flow across the range
First we examined how phenotype and population genetic structure vary across geographic distance. We used the standardized color traits to test whether color varies 107 across the range of Eurycea lucifuga. We used a nested linear model with the R function lm() to test whether traits varied among populations and among the major genetically differentiated clades in this species (see Chapter 2). Type II tests were performed with the Anova() function in the package car (Fox & Weisberg 2011). We also tested whether phenotypic differentiation varies across the range based on geographic distance with a
Mantel test, including pairwise PST as the dependent variable and geographic distance as the independent variable. We used the mantel() function in the R statistical package vegan (Oksanen et al. 2014) to perform Mantel tests, using the Pearson method with 1000 permutations. We next tested for isolation by distance by comparing pairwise FST with geographic distance using a linear model with the function lm(), as well as including FST and geographic distance in a Mantel test with the function mantel().
To compare phenotypic variation with neutral genetic variation, we compared estimates of pairwise differentiation across the range. For each color trait we compared phenotypic and genetic variation while controlling for geographic distance using a partial
Mantel test with the function mantel(). In each Mantel model we included pairwise PST as the x distance variable, pairwise FST as the y distance variable, and pairwise geographic distance as the z distance variable. By testing for correlations between PST and FST while controlling for geographic distance we were able to infer whether or not each color trait exhibits differentiation that is consistent with what we expect based on neutral evolutionary processes. If pairwise PST and pairwise FST are significantly correlated, we infer that similar neutral processes have influenced the evolution of both; however, a lack of significant correlation suggests that the pattern of differentiation in the color trait has been influenced by either diversifying or stabilizing selection depending on 108
the relationship of PST and FST (Antoniazza et al. 2010; Whitlock 2008). We also directly compared global PST and FST using student’s t-tests with the function t.test() in the R package to test whether overall there is greater phenotypic or neutral genetic divergence across the range.
Following this, we ran the same analyses on a smaller-scale dataset containing only populations from Mammoth Cave National Park to test whether the relationship between phenotypic and genetic differentiation changed at different spatial scales. The
Mammoth analyses used the same methods as we described previously.
Statistical analyses: Gene flow at small scales
We next used comparisons among genetic distance, geographic distance, and subterranean and surface connectivity to infer whether dispersal and/or gene flow occurs primarily over the surface or underground in Eurycea lucifuga. We used only the pairwise FST and pairwise geographic distances from the populations within Mammoth
Cave National Park (MCNP). We then manually populated a binary ‘cave connection’ matrix, indicating whether intersecting populations were connected or unconnected with each other via the main Mammoth Cave system. Lastly, we manually populated a binary
‘river connection’ matrix, indicating whether populations are on the same side
(‘connected’) or opposite sides (‘unconnected’) of the Green River, which runs through the park.
We first established whether our analyses may be biased by the arrangement of the cave entrances (e.g. if all cave entrances with subterranean connections were spatially clustered), so we tested the data for differences in pairwise geographic distance among connected and unconnected caves in a linear model. A second linear model tested the 109 effect of the two connectivity parameters (river and cave connectivity) on genetic distance, including geographic distance as a covariate. Both linear models were constructed in the function lm(), and Type II tests were implemented using the Anova() function. We then used Analysis of Molecular Variance in the software GenoDive with
999 permutations to examine how genetic variation was partitioned among the Mammoth
Cave National Park populations. One model was run including each population’s position relative to the Green River (North or South) as regional groups, and a second model included whether each population was connected to the Mammoth Cave system or not as groups. Finally, to better visualize relatedness among populations in the context of their connectedness below- and above-ground, we performed a Principal Coordinate
Analysis using the function pco() in the R package labdsv (Roberts 2013). This analysis projects measures of distance (in this case, genetic distance) in multiple dimensions, then reduces the dimensionality using Principal Components Analysis on the projection. The result is a visualization of genetic distance among cave entrance populations.
Results
Range-wide analysis
A visual summary of the means and standard errors of each principal component among the populations indicates that variation is greater in black spot PC1 than in either the orange background PC1 or the black:orange ratio (Figure 4). Most of this variation seems to exist within the central region, both in among- and within-population variance.
Results of an ANOVA testing for differences in color traits among regions indicate that there are no significant differences in orange PC1 or the black:orange ratio across the range of Eurycea lucifuga, but that black PC1 does vary significantly among 110 regions (Table 2). Mantel tests indicate that there is no significant correlation between pairwise geographic distance and orange PC1 (Mantel’s r=0.0335, p=0.335), black PC1
(Mantel’s r=0.062, p=0.238), or black:orange ratio (Mantel’s r=0.038, p=0.299). In contrast, we did find significant isolation by distance in a Mantel test including pairwise
FST and geographic distance (Mantel’s r=0.933, p<0.001).
We found no significant correlations between pairwise PST and pairwise FST when controlling for pairwise geographic distance in partial Mantel tests including orange PC1 or black PC1, which suggests that variation in these traits has not been structured by the neutral processes influencing neutral loci. Differentiation in orange PC1 increases faster across geographic distance than does FST, while differentiation in black PC1 increases slower than FST. However, the partial Mantel test for correlation between PST and FST including the black:orange ratio indicate a significant correlation between phenotypic and neutral genetic differentiation (Table 3; Figure 5). Significant differences between the global means of PST and FST for both orange PC1 and black PC1 but not black:orange ratio are consistent with the results of the Mantel tests (Table 4). A difference in the spatial structuring of variation in phenotypic traits and neutral genetic variation suggests that selection may be acting on both orange PC1 and black PC1. In contrast, these results suggest that color patterning, represented by the black:orange ratio, has been largely influenced by neutral processes.
Mammoth Cave National Park PST-FST comparisons
ANOVA results at this much smaller scale (0.35-10.69km) indicate that both black PC1 and black:orange ratio differed significantly among Mammoth populations
(Table 2). Tests for correlations between color trait differentiation and pairwise 111 geographic distance using Mantel tests were nonsignificant for orange PC1 (Mantel’s r=-
0.020, p=0.524), black PC1 (Mantel’s r=-0.138, p=0.799), and black:orange ratio
(Mantel’s r=-0.057, p=0.629). Likewise, there was no evidence of genetic isolation by distance among the Mammoth populations (Mantel’s r=-0.142, p=0.791).
When we performed partial Mantel tests to examine correlation between PST and
FST while accounting for geographic distance, we found that there were no significant correlations for any of the three traits, although the relationship between black:orange ratio and FST was approaching significance (Table 3). However, t-tests comparing the means of PST and FST for each color trait indicated that all three traits differed significantly from mean neutral genetic differentiation, with PST greater than FST in each case (Table 4).
Sensitivity analyses
Our sensitivity analyses indicate that heritability only moderately affects the estimation of PST. However, different values of g (the additive genetic variation among populations) strongly influence the magnitude of PST and therefore its relationship with
FST (Fig. 6, Table S2). It is possible that if g is smaller than the estimate we chose to work with, PST may be significantly lower than FST in all traits and at all spatial scales.
This would suggest that stabilizing selection has acted on color. Because of our inability to accurately predict the correct estimates of h2 and g, we are able to make better predictions about the difference in spatial structuring of phenotypic and neutral genetic variation than about direct comparisons between the two estimates of differentiation. 112
Mammoth Cave National Park genetic and geological comparisons
There was no significant difference in geographic distance among caves connected as part of the Mammoth Cave System versus unconnected caves (F=1.814;
DF=1,180; p=0.18), indicating that genetic differentiation among these two groups of caves was not due to differences in the spatial distribution of cave entrances. Whether or not two cave entrances are connected within the Mammoth Cave system had a significant impact on the genetic distance between those populations (F=11.391; DF=1,179; p<0.001). Unexpectedly, populations connected underground through the Mammoth
Cave complex exhibited greater genetic differentiation than populations outside the
Mammoth complex, indicating that underground connections present a barrier to gene flow relative to surface distance. Finally, populations on the same side of the Green
River were significantly less genetically differentiated than those on opposing sides of the river (F=4.131, DF=1,178, p=0.044), demonstrating the Green River has been a significant migratory barrier for this species (Figure 7).
Estimates of population structure indicate that the vast majority of variance in
Mammoth Cave National Park is contained within individuals (approximately 84%), and most of the rest of the genetic variance is partitioned among individuals within populations (approximately 13%). Very little variance was partitioned among groups when we included either connectedness with the Mammoth cave system or via the Green
River in the AMOVA model (Table 5). These patterns of genetic isolation can be detected visually using Principal Coordinate Analysis to cluster populations (Figure 8).
Populations on either side of the Green River were similarly genetically variable (i.e. they spanned a similar broad range of both PCo Axes), but for populations south of the Green 113
River the locations along the two PCo Axes co-varied in a positive way whereas the opposite was true for populations North of the Green River which likely accounts for the genetic differentiation between the two groups. South of the Green River, populations collected from Mammoth Cave spanned the entire range of genetic variation on both axes, whereas populations not connected to the Mammoth Cave complex had less genetic variance along the first PCo Axis. (Figure 8).
Taken together, significant isolation by distance at larger spatial scales, genetic differentiation across the Green River, and genetic differentiation among cave populations known to be interconnected by subterranean aquatic corridors suggest migration is achieved primarily by surface terrestrial routes in this species.
Discussion
We compared differentiation in color traits with that of neutral genetic variation in
Eurycea lucifuga at different spatial scales in order to assess the selective maintenance of warning coloration in a cave-dwelling salamander. We found that most of the variance in the orange background among salamanders is accounted for by variation in the saturation and value of the orange, and not the hue itself. However, hue, saturation, and value all contributed equally to the variance seen in the black spot trait. We found significant differences in black spot traits across the range, but not the orange background nor the ratio of spot to background. Additionally, there were several indications that non-neutral processes have affected some of these traits: First, we saw that differentiation in the orange and black traits exhibited different spatial structuring than neutral genetic differentiation. Second, while the black spot trait exhibited no isolation by distance within the Mammoth Cave National Park, and differentiation was not correlated with that 114 of neutral genetic variation, there were significant differences in black spots among
Mammoth populations, which suggests that the structure of variation is influenced by processes other than those influencing the neutral genetic differentiation. Third, our comparison of genetic differentiation among Mammoth populations with subterranean and surface barriers provided a potential explanation for non-neutral influences on color traits: that dispersal most likely occurs across the surface habitat in this species rather than through cave conduits. However, it is important to note that our interpretations of whether color is under diversifying or stabilizing selection depends on the additive genetic variance in these traits, which is unknown.
We also found differences in the spatial structure of the different color traits, suggesting that they are influenced by different selective or neutral forces. Patterns of trait variation differ between the orange background and black spot traits and the black to orange ratio, with the orange background and black to orange ratio exhibiting smaller differences among means as well as smaller standard errors than the black spots.
Variation among populations in black spots is especially notable among central populations, which is the most variable of the regions. Comparisons of the spatial structuring of variation between these color traits and neutral genetic differentiation suggests that both the black and orange color traits have been influenced by non-neutral processes, but that the ratio of black to orange may have experienced a relaxation of selection and has been influenced by neutral processes.
Sensitivity analyses indicate that our estimates of PST depend on the amount of additive genetic variance we included in our models for each trait. While we would not expect to see a change in the spatial relationships between FST and PST over distance, a 115
change in additive genetic variance in color traits would alter the magnitude of PST for each trait. This would alter our interpretations of how selection may be acting on these traits. For example, though phenotypic differentiation in the orange hue is greater at all scales than neutral genetic differentiation, as described above, if additive genetic variance plays a smaller role in the variance of that trait it may change the relationship and suggest a neutral process or purifying selection. While our results support a role of possible selective maintenance of color traits, without a better estimation of both heritability and additive genetic variance we are better able to make inferences about the differences in spatial structure between PST and FST than about the role of different selective forces.
The results of this study reinforce a suggestion made by others that the scale at which a study is conducted has important effects on its results and implications in studies of both population and ecological dynamics as well as selection analyses (Addicott et al.
1987; Barton & Slatkin 1986; Levene 1953; Svensson & Sinervo 2004). For example,
Svensson and Sinervo (2004) found that variation in local selective regimes was undetected when data were analyzed on a global scale, which suggests that not only should scale be an important parameter when designing studies, taking into account aspects of a species’ ecology such as dispersal and predation, but that analysis at multiple scales is a valuable tool for understanding selection regimes. In our case, had we restricted our comparison of phenotypic and neutral genetic variation to a local scale we would have observed that phenotypic differentiation is great than neutral genetic differentiation and hypothesized that this was due to local adaptation, but ignored the important role that spatial structuring plays in variation of color traits in this species. 116
However, by examining color traits and neutral genetic loci at both range-wide and local scales, we were able to capture the complicated relationships among their variation.
There are many possible explanations for why and how spatial structuring of color variation may differ from neutral expectations in this system. First, our finding that fine- scale genetic differentiation is influenced more by surface features than cave features suggests that movement across the surface may expose individuals to environmental and community dynamics that make particular color patterns beneficial as visual cues. A potential role for the bright coloration in Eurycea lucifuga is aposemetism or mimicry, as is seen in other ‘red salamander’ systems (Howard & Brodie 1971; Kuchta & Reeder
2005; Tilley et al. 1982). Salamanders are generally profitable prey for animals like birds, mammals, snakes, fishes, turtles, frogs, crayfish, predatory insects, and other salamanders (Petranka 1998), and the selective pressure imposed by predators could account for the lack of variation in the orange background trait at both rangewide and small scales, as well as its differential spatial relationship with neutral genetic variation.
It is particularly suggestive that the hue of the orange background is particularly invariant compared to the saturation and value components of background color. Predation pressure during dispersal across the surface habitat plausibly explains this lack of variation. Though we cannot address specific correlations between the variation we found in color traits and the ability of predators to perceive those differences, it has been shown in birds that red coloration promotes predator avoidance (Brodie Jr & Brodie III
1980; Tilley et al. 1982), but that patterning does not impact a predator’s ability to learn avoidance (Aronsson & Gamberale-Stille 2008). This may explain why we see potentially evidence of selection on orange color but not spot patterning. 117
Selection from predator pressure is by no means the only potential explanation for maintenance of color in Eurycea lucifuga. Color has also been found to play physiological roles that may drive spatial variation, as in the thermoregulatory properties underlying feather color in bearded vultures Gypaetus barbatus barbatus (Margalida et al. 2008). In many species coloration is reflective of individual condition. For example, different color morphs in the wall lizard, Podarcis muralis, exhibit correlated differences in body size, parasite prevalence and infection intensity, running stamina, and immune function (Calsbeek et al. 2010). Similarly, color variation has been found to be an honest signal of parasitism, which influences mate choice in fence lizards, Sceloperus occidentalis (Ressel & Schall 1989), and of general body condition in house finches,
Carpodacus mexicanus (Hill 2000). Color pattern variation rather than color variation can also indicate poor condition, as in the correlation between asymmetrical spot numbers and body asymmetricality in the spotted salamander, Ambystoma maculatum
(Davis & Maerz 07). Coloration has also been shown to be an important component of mate attraction and courtship, especially in species with a visual courtship display such as red-spotted newts, Notophthalmus v. viridescens (Davis & Grayson 2008) and the ruff,
Philomachus pugnax (Widemo 1998). There can also be large effects of ontogeny and development on coloration and color patterning, as McClureand McCune (2003) found in zebrafish, in which pigment patterning was influenced by changes in size and shape during growth. Arizona tiger salamanders, Ambystoma tigrinum, experience change in coloration over time, growing paler as they age (Fernandez & Collins 1988). Finally, neutral genetic processes could explain the differences in spatial structure of phenotype and genotype we see, through processes such as bottleneck events during range 118 expansion, or a homogenizing effect of local differentiation across regions. Direct measurements of selection in this species will help us better understand which processes are influencing color traits in Eurycea lucifuga.
The results of our examination of dispersal within Mammoth Cave National Park suggest migration occurs primarily via surface dispersal, but the patterns of dispersal are likely to be more complex. The greater genetic distance between caves connected by the
Mammoth Cave system is primarily the result of three populations (Historical, New
Discovery and Violet City) that span nearly all the range of genetic divergence across
PCoA1 and PCoA2 despite being relatively close geographically (Figure 8). We hypothesize that the historical dispersal among these cave entrances are likely to be a complex mix of colonization, extinction and founder effects. These populations are also embedded within a ridge and valley system that may drive patterns of land-based dispersal.
Finally, our results offer a striking contrast with a similar European terrestrial troglophilic salamander genus, Hydromantes, which exhibits very little dispersal among caves, and for whom the surface habitat is clearly a barrier (Chiari et al. 2012).
Troglophilic cricket populations also exhibit an avoidance of surface habitats, and restrictedness to cave habitats (Caccone & Sbordoni 1987). Invertebrate troglophiles have also shown variance in their dispersal capabilities, resulting from both environmental variation as well as intrinsic phenotypic traits (Caccone 1985).
Terrestriality has not proved to be a barrier to subterranean dispersal in other cave- dwelling species in the Mammoth Cave system. The cave beetle, Ptomophagus hirtus, for example, exhibits frequent dispersal across the Green River, which has led to 119 hypotheses that dispersal occurs via hydrological conduits underground despite this species’ typical avoidance of aquatic habitats (Laing et al. 1976). High genetic uniformity among populations has also been found in other invertebrate species that typically avoid the surface habitat, the cave cricket Hadenoecus subterraneus (Caccone
& Sbordoni 1987) and the Kentucky cave beetle, Neaphaenops tellkampfii (Giuseffi et al.
1978). However, dispersal abilities of the vertebrate inhabitants of Mammoth Cave have not been examined. Many of the unique aspects of Eurycea lucifuga’s biology are likely the result of the fact that despite being a cave inhabitant, subterranean aquatic habitats rather than surface habitats present the most significant barriers to dispersal. It is interesting to note that many of our generalities about adaptation to the cave environment stem from aquatic species and invertebrates that might follow a very different course of dispersal and adaptive evolution after moving underground.
Acknowledgements
The authors would like to acknowledge the contributions of G. Moni, T. Loring, J.
Lewis, K. Dunlap, D. Everton, J. Pearson, S. Hammond, C. Evans, M. Futrell, A. Futrell,
W. Orndorff, R. Stewart, R. Toomey, R. Olson, B. Balfour, and many other caving enthusiasts and wildlife professionals at each sampling locality in the collection of samples. Collections were made with the permission of the Missouri Department, West
Virginia Division of Natural Resources, Kentucky Department of Fish and Wildlife,
United States Department of the Interior National Park Service Mammoth Cave National
Park, Indiana Department of Natural Resources Division of Fish and Wildlife, Oklahoma
Department of Wildlife Conservation, The Nature Conservancy (Mill Creek Cave), 120
Virginia Department of Game and Inland Fisheries, and the Virginia Department of
Conservation and Recreation. R code for Fst-Pst analyses was provided by Andrea
Berardi and Malcolm Augat. Digital images for color measurements were processed by
Daniel Foley (UVA). Additionally, we would like to thank Andrea Berardi for her thoughtful comments on this manuscript.
121
Works Cited
(R Development Core Team, 2008) R: A language and environment for statistical
computing, R Foundation for Statistical Computing, Vienna, Austria.
Addicott JF, Aho JM, Antolin MF, et al. (1987) Ecological Neighborhoods: Scaling
Environmental Patterns. Oikos 49, 340-346.
Alho J, Herczeg G, Soderman F, et al. (2010) Increasing melanism along a latitudinal
gradient in a widespread amphibian: local adaptation, ontogenic or
environmental plasticity? BMC Evol Biol 10, 2-9.
Antoniazza S, Burri R, Fumagalli L, Goudet J, Roulin A (2010) Local adaptation
maintains clinal variation in melanin-based coloration of european barn owls
(Tyto alba). Evolution 64, 1944-1954.
Aronsson M, Gamberale-Stille G (2008) Domestic chicks primarily attend to colour,
not pattern, when learning an aposematic coloration. Animal Behaviour 75,
417-423.
Avise JC, Selander RK (1972) Evolutionary Genetics of Cave-Dwelling Fishes of the
Genus Astyanax. Evolution 26, 1-19.
Barton NH, Slatkin M (1986) A Quasi-equilibrium theory of the distribution of rare
alleles in a subdivided population. Heredity 56, 409-415.
Berardi AE, Fields PD, Abbate JL, Taylor DR (In review) Elevational divergence in
floral color in Silene vulgaris (Caryophyllacaea).
Bonett RM, Steffen MA, Lambert SM, Wiens JJ, Chippindale PT (2014) Evolution of
paedomorphosis in plethodontid salamanders: ecological correlates and re-
evaluation of metamorphosis. Evolution 68, 466-482. 122
Brandon RA, Huheey JE (1975) Diurnal Activity, Avian Predation, and the Question
of Warning Coloration and Cryptic Coloration in Salamanders. Herpetologica
31, 252-255.
Brandon RA, Labanick GM, Huheey JE (1979) Relative Palatability, Defensive
Behavior, and Mimetic Relationships of Red Salamanders (Pseudotriton
ruber), Mud Salamanders (Pseudotriton montanus), and Red Efts
(Notophthalmus viridescens). Herpetologica 35, 289-303.
Brodie ED, Jr. (1977) Salamander Antipredator Postures. Copeia 1977, 523-535.
Brodie ED, Jr. (1981) Phenological Relationships of Model and Mimic Salamanders.
Evolution 35, 988-994.
Brodie ED, Jr., Nowak RT, Harvey WR (1979) The Effectiveness of Antipredator
Secretions and Behavior of Selected Salamanders against Shrews. Copeia
1979, 270-274.
Brodie Jr ED, Brodie III ED (1980) Differential avoidance of mimetic salamanders by
free-ranging birds. Science 208, 181-182.
Caccone A (1985) Gene Flow in Cave Arthropods: A Qualitative and Quantitative
Approach. Evolution 39, 1223-1235.
Caccone A, Sbordoni V (1987) Molecular Evolutionary Divergence Among North
American Cave Crickets. I. Allozyme Variation. Evolution 41, 1198-1214.
Calsbeek B, Hasselquist D, Clobert J (2010) Multivariate phenotypes and the
potential for alternative phenotypic optima in wall lizard (Podarcis muralis)
ventral colour morphs. J Evol Biol 23, 1138-1147. 123
Chiari Y, van der Meijden A, Mucedda M, et al. (2012) Phylogeography of Sardinian
Cave Salamanders (Genus Hydromantes) Is Mainly Determined by
Geomorphology. PLoS ONE 7, e32332.
Culver D, Master L, Christman M (2000) Obligate cave fauna of the 48 contiguous
United States. Conservation Biology 14, 386-401.
Davis AK, Grayson K, L. (2008) Spots of adult male red-spotted newts are redder and
brighter than in females: evidence for a role in mate selection? Herpetological
Journal 18, 83-89.
Davis AK, Maerz JC (07) Spot symmetry predicts body condition in spotted
salamanders, Ambystoma maculatum. Applied Herpetology 4, 195-205.
Fernandez PJ, Jr., Collins JP (1988) Effect of Environment and Ontogeny on Color
Pattern Variation in Arizona Tiger Salamanders (Ambystoma tigrinum
nebulosum Hallowell). Copeia 1988, 928-938.
Ficetola G, Pennati R, Manenti R (2012) Spatial segregation among age classes in
cave salamanders: habitat selection or social interactions? Population
Ecology, 1-10.
Ficetola G, Pennati R, Manenti R (2013) Spatial segregation among age classes in
cave salamanders: habitat selection or social interactions? Population
Ecology 55, 217-226.
Fox J, Weisberg S (2011) An {R} Companion to Applied Regression, Second Edition
Sage, Thousand Oaks, CA. 124
Giuseffi S, Kane TC, Duggleby WF (1978) Genetic Variability in the Kentucky Cave
Beetle Neaphaenops tellkampfii (Coleoptera: Carabidae). Evolution 32, 679-
681.
Hansen TF, Pélabon C, Houle D (2011) Heritability is not Evolvability. Evolutionary
Biology 38, 258-277.
Hill GE (2000) Energetic Constraints on Expression of Carotenoid-Based Plumage
Coloration. Journal of Avian Biology 31, 559-566.
Hoffman EA, Blouin MS (2004) Evolutionary history of the Northern Leopard Frog:
reconstruction of phylogeny, phylogeography, and historical changes in
population demography from mitochondrial DNA. Evolution 58, 145-159.
Howard RR, Brodie ED (1971) Experimental Study of Mimicry in Salamanders
involving Notophthalmus viridescens viridescens and Pseudotriton ruber
schencki. Nature 233, 277-277.
Howard RR, Brodie ED, Jr. (1973) A Batesian Mimetic Complex in Salamanders:
Responses of Avian Predators. Herpetologica 29, 33-41.
Hutchison V (1958) The Distribution and Ecology of the Cave Salamander, Eurycea
lucifuga. Ecological Monographs 28, 1-20.
Jaeger RG (1981) Birds as Inefficient Predators on Terrestrial Salamanders. The
American Naturalist 117, 835-837.
Jeffery W, Strickler A, Yamamoto Y (2003) To see or not to see: evolution of eye
degeneration in Mexican blind cavefish. Integrative Comparative Biology 43,
531-541. 125
Jeffery WR (2009) Regressive Evolution in Astyanax Cavefish. Annual Review of
Genetics 43, 25-47.
Jones R, Culver DC, Kane TC (1992) Are Parallel Morphologies of Cave Organisms
the Result of Similar Selection Pressures? Evolution 46, 353-365.
Kats LB, Petranka JW, Sih A (1988) Antipredator Defenses and the Persistence of
Amphibian Larvae With Fishes. Ecology 69, 1865-1870.
Kraemer AC, Adams DC (2014) Predator perception of Batesian mimicry and
conspicuousness in a salamander. Evolution 68, 1197-1206.
Kuchta SR, Reeder TW (2005) Experimental Support for Aposematic Coloration in
the Salamander Ensatina eschscholtzii xanthoptica: Implications for Mimicry
of Pacific Newts. Copeia 2005, 265-271.
Laing C, Carmody GR, Peck SB (1976) Population Genetics and Evolutionary Biology
of the Cave Beetle Ptomaphagus hirtus. Evolution 30, 484-498.
Leinonen T, Cano JM, Mäkinen H, Merilä J (2006) Contrasting patterns of body shape
and neutral genetic divergence in marine and lake populations of threespine
sticklebacks. Journal of Evolutionary Biology 19, 1803-1812.
Levene H (1953) Genetic Equilibrium When More Than One Ecological Niche is
Available. The American Naturalist 87, 331-333.
Lindström L (1999) Experimental Approaches to Studying the Initial Evolution of
Conspicuous Aposematic Signalling. Evolutionary Ecology 13, 605-618.
Lunghi E, Manenti R, Ficetola GF (2014) Do cave features affect underground habitat
exploitation by non-troglobite species? Acta Oecologica 55, 29-35. 126
Manier MK, Seyler CM, Arnold SJ (2007) Adaptive divergence within and between
ecotypes of the terrestrial garter snake, Thamnophis elegans, assessed with
FST-QST comparisons. Journal of Evolutionary Biology 20, 1705-1719.
Margalida A, Negro JJ, Galván I (2008) Melanin-based color variation in the Bearded
Vulture suggests a thermoregulatory function. Comparative Biochemistry and
Physiology Part A: Molecular & Integrative Physiology 149, 87-91.
McClure M, McCune AR (2003) Evidence for developmental linkage of pigment
patterns with body size and shape in Danios (Teleostei: Cyprinidae).
Evolution 57, 1863-1875.
Meirmans PG, Van Tienderen PH (2004) GENOTYPE and GENODIVE: two programs
for the analysis of genetic diversity of asexual organisms. Molecular Ecology
Notes 4, 792-794.
Merilä J, Crnokrak P (2001) Comparison of genetic differentiation at marker loci and
quantitative traits. Journal of Evolutionary Biology 14, 892-903.
Oksanen J, Blanchet FG, Kindt R, et al. (2014) vegan: Community ecology package. R
package version 2.2-0.
Petranka J (1998) Salamanders of the United States and Canada Smithsonian
Institution Press, Washington.
Pipan T, Culver DC (2012) Convergence and divergence in the subterranean realm: a
reassessment. Biological Journal of the Linnean Society.
Porter M (2007) Subterranean biogeography: What have we learned from molecular
techniques? Journal of Cave and Karst Studies 69, 179-186. 127
Poulton EB (1890) The colours of animals: their meaning and use, especially
considered in the case of insects D. Appleton.
Protas M, Conrad M, Gross JB, Tabin C, Borowsky R (2007) Regressive Evolution in
the Mexican Cave Tetra, Astyanax mexicanus. Current Biology 17, 452-454.
Rasband WS (1997-2014) ImageJ. National Institutes of Health, Bethesda, Maryland.
Ressel S, Schall JJ (1989) Parasites and Showy Males: Malarial Infection and Color
Variation in Fence Lizards. Oecologia 78, 158-164.
Roberts DW (2013) Ordination and multivariate analysis for ecology.
Rudh A, Qvarnström A (2013) Adaptive colouration in amphibians. Seminars in Cell
& Developmental Biology 24, 553-561.
Sæther SA, Fiske P, Kålås JA, et al. (2007) Inferring local adaptation from QST–FST
comparisons: neutral genetic and quantitative trait variation in European
populations of great snipe. Journal of Evolutionary Biology 20, 1563-1576.
Sambrook J, Fritsch E, T M (1989) Molecular Cloning: A laboratory Manual Cold
Springs Harbor Laboratory Press, New York.
Schlegel PA, Steinfartz S, Bulog B (2009) Non-visual sensory physiology and
magnetic orientation in the Blind Cave Salamander, Proteus anguinus (and
some other cave-dwelling urodele species). Review and new results on light-
sensitivity and non-visual orientation in subterranean urodeles (Amphibia).
Animal Biology 59, 351-384.
Sket B (2008) Can we agree on an ecological classification of subterranean animals?
Journal of Natural History 42, 1549-1563. 128
Spitze K (1993) Population structure in Daphnia obtusa: quantitative genetic and
allozymic variation. Genetics 135, 367-374.
Storz JF (2002) Contrasting patterns of divergence in quantitative traits and neutral
DNA markers: analysis of clinal variation. Molecular Ecology 11, 2537-2551.
Strecker U, Faúndez VH, Wilkens H (2004) Phylogeography of surface and cave
Astyanax (Teleostei) from Central and North America based on cytochrome b
sequence data. Molecular phylogenetics and evolution 33, 469-481.
Svensson EI, Sinervo B (2004) Spatial Scale and Temporal Component of Selection in
Side‐Blotched Lizards. The American Naturalist 163, 726-734.
Tilley SG, Lundrigan BL, Brower LP (1982) Erythrism and Mimicry in the
Salamander Plethodon cinereus. Herpetologica 38, 409-417.
Vavrek MJ (2011) fossil: paleoecological and paleogeographical analysis tools.
Palaeontologia Electronica 14.
Walsh PS, Metzger DA, Higuchi R (1991) Chelex 100 as a medium for simple
extraction of DNA for PCR-based typing from forensic material. Biotechniques
10, 506-513.
Wessel A, Erbe P, Hoch H (2007) Pattern and process: Evolution of troglomorphy in
the cave-planthoppers of Australia and Hawai'i–Preliminary observations
(Insecta: Hemiptera: Fulgoromorpha: Cixiidae). Acta Carsologica 36, 199-
206.
Whitlock MC (2008) Evolutionary inference from QST. Molecular Ecology 17, 1885-
1896. 129
Widemo F (1998) Alternative reproductive strategies in the ruff, Philomachus
pugnax: a mixed ESS? Animal Behaviour 56, 329-336.
Wilson RE, Peters JL, McCracken KG (2013) Genotypic and phenotypic divergence
between low- and high-altitude populations of two recently diverged
Cinnomon Teal subspecies. Evolution 67, 170-184.
Yamamoto Y, Stock DW, Jeffery WR (2004) Hedgehog signalling controls eye
degeneration in blind cavefish. Nature 431, 844-847.
Zhang Y, Li S (2013) Ancient lineage, young troglobites: recent colonization of caves
by Nesticella spiders. BMC Evol Biol 13, 183.
Zhou M, Johnson AM, Fuller RC (2014) Patterns of Male Breeding Color Variation
Differ across Species, Populations, and Body Size in Rainbow and
Orangethroat Darters. Copeia 2014, 297-308.
130
Figures:
Figure 1. Adult Eurycea lucifuga, on its most common substrate, the rock wall of caves.
Figure 2. Collection locations (black markers) superimposed over the range of Eurycea lucifuga (grey).
Size of locality markers is representative of the number of samples collected from that locality (ranging from 1 to 11).
131
North(of(Green( Phil( River( Hickory( South(of(Green( Great(Onyx( River( 37.22 Big(Hollow( Connected(with( Mammoth( Isolated(from( Crystal( Mammoth(
37.20 Falling(Tree( Longitude Historical( C2( Cadaverous( White( 37.18 YMCA( New(Discovery(
Violet(City(
−86.125 −86.100 −86.075 −86.050 Latitude
Figure 3. Geographic distribution of collecting localities within the local area of Mammoth Cave National
Park. Colors and shapes represent connection underground via the Mammoth Cave system, and aboveground in relation to the Green River, as shown in the legend.
132
Central( Indiana( Western( 0.10
0.05 Transect
C 0.00 I
−0.05 W Orange PC1 Orange
−0.10
−0.15
1.2
C2 Big Bull Eoff Iron Jail Lost Phil Blue Great New Stans Third Violet White Austen CrewsCrystal Falling YMCA Gillespie Hickory January Sullivan Historical Survivalist Cadaverous 0.8 Population Transect
C 0.4 I
W Black PC1
0.0
−0.4
0.5 C2 Big Bull Eoff Iron Jail Lost Phil Blue Great New Stans Third Violet White Austen CrewsCrystal Falling YMCA Gillespie Hickory January Sullivan Historical Survivalist Cadaverous 0.4 Population Transect
C 0.3 I
0.2 W
Orange:Black Ratio Orange:Black 0.1
C2 Big Bull Eoff Iron Jail Lost Phil Blue Great New Stans Third Violet White Austen CrewsCrystal Falling YMCA Gillespie Hickory January Sullivan Historical Survivalist Cadaverous Population
Figure 4. Means and standard errors of each color trait within each population across the range. Shades indicate regions within the range of Eurycea lucifuga- black=central, white=Indiana, grey=Western.
133
Orange+PC1+ Black+PC1+ Black:Orange+Ra5o+ P + F + 1.0 1.0 1.0 ST ST 0.8 0.8 0.8 0.6 0.6 0.6 0.4 0.4 0.4 Pairwise Differentiation Pairwise Differentiation Pairwise Differentiation Pairwise Differen5a5on+ 0.2 0.2 0.2 0.0 0.0 0.0
0 200 400 600 800 0 200 400 600 800 0 200 400 600 800 Pairwise Distance (km) Pairwise Distance (km) Pairwise Distance (km) Geographic+Distance+(km)+
Figure 5. Pairwise PST values (dotted line, open circles) for orange background PC1, black spot PC1, and black:orange ratio, each in comparison with pairwise FST (black line, filled circles), plotted against pairwise
0.5 geographic distance in kilometers. "0.5"
g=1 variable g=.1
g=.01 Fst
g = 1 Legend g = 0.1 g = 0.01
Fst Sensitivity Analysis
FST* g=1* g=0.1* g=0.01*1.00
0.4 SensitivityOrange* Analysis SensitivityBlack* Analysis Orange:BlackSensitivity Analysis*Ra=o* 0.75 0.4 0.3 0.5
0.4 0.50
0.2 "0.5" "0.5" "0.5" Index Heritability 0.5 0.5 0.5 0.4 0.3 0.25 0.1 variable variable variable
0.3 g=1 g=1 g=1
g=.1 0.0 0.00 g=.1 g=.1 0.3 g=.01 g=.01 g=.01 0.2 Fst Fst Fst Index Index 0.2 Index Legend Index* Legend Legend 0.2 g = 1 g = 1 g = 1
g = 0.1 g = 0.1 g = 0.1 0.1 0.1 g = 0.01 g = 0.01 g = 0.01
0.1 Fst Fst Fst
0.0 0.0
0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00 Heritability Heritability Heritability Heritability*
Figure 6. Results of a sensitivity analysis in which mean PST was estimated for each color trait using multiple values of heritability (0, 0.25, 0.5, 0.75, and 1) and additive genetic variance (0.01, 0.1, and 1).
These are plotted in comparison with FST (black line).
134
*"
0.09 0.09
0.06 0.06 Genetic Distance Genetic Distance 0.03 0.03
0.00 0.00 Mammoth Isolated Connected Unconnected Cave River Figure 7. A comparison of pairwise FST between caves which are either connected or disconnected in the context of subterranean paths (Cave) and the Green River (River). Bars represent 95% confidence intervals.
North*of*Green* Phil* River* Historical* Hickory* South*of*Green* Hickory* Great*Onyx* River* 37.22 Big*Hollow* Connected*with* Mammoth* 0.02 YMCA* Isolated*from* Crystal* Mammoth*
Cadaverous* Big*Hollow* 37.20 Falling*Tree* New*Discovery* 0.00 PCO2 Falling*Tree* Longitude Historical* C2* Crystal* Cadaverous* C2* Great*Onyx* White* White* 37.18 YMCA* New*Discovery* −0.02 Violet*City* Violet*City* Phil*
−86.125 −86.100 −86.075 −86.050 −0.06 −0.03 0.00 0.03 0.06 Latitude PCO1
Figure 8. A comparison of cave entrances plotted according to their latitude and longitude (left), and according to their genetic similarity as estimated using the first two axes of a Principal Coordinate Analysis
(PCoA; right).
Tables:
135
Table 1. Results of Principal Components Analyses performed on the rangewide data and the Mammoth
Cave National Park data using hue, saturation, and value for both the orange background and black spot traits.
Rangewide Orange Background
PC1 PC2 PC3
Standard Deviation 0.064 0.021 0.009
Prop. of Variance 0.886 0.098 0.016
Cumulative Variance 0.886 0.984 1.000
Hue -0.113 -0.928 0.355
Saturation -0.505 -0.254 -0.825
Value -0.856 0.272 0.440
Black Spot
PC1 PC2 PC3
Standard Deviation 0.444 0.284 0.159
Prop. of Variance 0.651 0.266 0.083
Cumulative Variance 0.651 0.917 1.000
Hue 0.520 0.786 0.334
Saturation 0.623 -0.082 -0.778
Value 0.584 -0.612 0.533
Mammoth Orange Background
PC1 PC2 PC3
Standard Deviation 0.091 0.015 0.010
Prop. of Variance 0.962 0.026 0.012
Cumulative Variance 0.962 0.988 1.000
Hue 0.058 0.955 -0.291
Saturation 0.486 0.228 0.844
Value 0.872 -0.190 -0.451
Black Spot
PC1 PC2 PC3
Standard Deviation 0.508 0.385 0.184
Prop. of Variance 0.587 0.337 0.077
Cumulative Variance 0.587 0.923 1.000
Hue 0.514 -0.810 0.283
Saturation 0.627 0.130 -0.768
Value 0.585 0.572 0.574
Table 2. ANOVA results depicting color trait differences among geographic regions across the range including Region as a fixed effect and Population nested within region as a random effect, and among 136 populations within Mammoth Cave National Park including Population as the fixed effect. Significant p values are in bold.
Trait Chi-square DF p
Range-wide Orange PC1 4.185 2, 147 0.123
Black PC1 10.844 2, 147 0.004
Black:Orange Ratio 0.143 2, 147 0.931
Trait SS DF F p
Mammoth Orange PC1 0.087 12, 55 0.866 0.585 Black PC1 7.486 12, 55 3.495 0.001
Black:Orange Ratio 0.173 12, 55 2.398 0.014
Table 3. Mantel tests, conducted using 999 permutations, included a comparison of range-wide Pst by geographic distance and PST-FST by geographic distance. Significant correlations are in bold.
Range-wide Mammoth
Trait Mantel's r p Mantel's r p Orange PC1 0.01 0.449 0.19 0.123 Black PC1 -0.108 0.872 0.07 0.333 Black:Orange Ratio 0.258 0.005 0.223 0.073
Table 4. Results of independent and pairwise student’s t-tests comparing rangewide PST of each color trait to pairwise FST, which had a mean of 0.2299.
t DF p
Range-wide Orange PC1 2.076 1292 0.038 Black PC1 6.847 1176 <0.0001
Black:Orange Ratio -1.813 1389 0.07
Mammoth Orange PC1 8.577 159 <0.0001 Black PC1 11.496 157 <0.0001
Black:Orange Ratio 10.099 158 <0.0001
137
Table 5. Results of two Analyses of Molecular Variance, including both the side of the Green River that a cave entrance is located, and whether or not it is connected with the main Mammoth Cave system as regional partitions. Significant p-values are in bold.
Source of Variation Nested in %var F-stat F-value p Within Individual -- 0.832 R_it 0.168 -- Among Individual Population 0.124 R_is 0.129 0.018 Among Population N or S of River 0.028 R_sc 0.029 0.126 Among River Groups -- 0.016 R_ct 0.016 0.181
Source of Variation Nested in %var F-stat F-value p Within Individual -- 0.841 R_it 0.159 -- Among Individual Population 0.13 R_is 0.134 0.012 Among Population Cave/Isolated 0.046 R_sc 0.045 0.048 Among Cave/Isolated -- -0.017 R_ct -0.017 0.789
138
Supplementary Tables:
S1: Collecting information and raw color values for Hue, Saturation, and Value of each individual’s orange background, black spots, and the standard to which each was compared. Spot Ratio is the ratio of spot area to background area.
Orange Background Black Spot Standard Region ID Population Hue Saturation Value Hue Saturation Value Hue Saturation Value Spot Ratio 1 Central Aus6.30.1 Austen 0.0669 0.0587 0.0357 0.4941 0.0211 0.0124 0.0939 0.0705 0.0568 0.2821 2 Central Aus7.2.1 Austen 0.0357 0.0542 0.0675 0.0124 0.0301 0.0382 0.0568 0.0784 0.0869 0.2173 3 Central Big7.3.1 Big Hollow 0.0371 0.0406 0.0350 0.0101 0.0097 0.0131 0.0422 0.0457 0.0500 0.1960 4 Central Big7.3.2 Big Hollow 0.0350 0.0268 0.0251 0.0131 0.5012 0.4963 0.0500 0.0482 0.0494 0.4184 5 Central Big7.3.3 Big Hollow 0.0251 0.0352 0.0474 0.4963 0.0102 0.0140 0.0494 0.0519 0.0476 0.2885 6 Central Big7.3.4 Big Hollow 0.0474 0.0407 0.0336 0.0140 0.0119 0.0098 0.0476 0.0421 0.0419 0.3961 7 Central Big7.3.5 Big Hollow 0.0336 0.0298 0.0281 0.0098 0.0065 0.4894 0.0419 0.0469 0.0493 0.3768 8 Central Big7.3.6 Big Hollow 0.0281 0.0268 0.0250 0.4894 0.4932 0.0096 0.0493 0.0482 0.0475 0.2632 9 Central Big7.3.7 Big Hollow 0.0250 0.0396 0.0486 0.0096 0.0184 0.4756 0.0475 0.5171 0.9896 0.2058 10 Central C27.1.1 Frank's Cave 0.0209 0.0326 0.0492 0.4965 0.4924 0.4938 0.0454 0.0484 0.0516 0.1779 11 Central C27.1.2 Frank's Cave 0.0492 0.0458 0.0423 0.4938 0.0057 0.0071 0.0516 0.0504 0.0491 0.1832 12 Central C27.1.3 Frank's Cave 0.0423 0.0371 0.0516 0.0071 0.4832 0.4881 0.0491 0.0479 0.0685 0.3511 13 Central C27.1.4 Frank's Cave 0.0305 0.0727 0.0650 0.9590 0.0172 0.0044 0.0465 0.0905 0.0941 0.3452 14 Central C27.2.1 Frank's Cave 0.0689 0.0666 0.0663 0.0108 0.0051 0.0074 0.0923 0.0782 0.0571 0.1896 15 Central C27.2.2 Frank's Cave 0.0663 0.0688 0.0671 0.0074 0.0142 0.4843 0.0571 0.0692 0.0794 0.2266 16 Central C27.2.3 Frank's Cave 0.0732 0.0610 0.0617 0.0194 0.9492 0.9799 0.0865 0.0723 0.0915 0.1887 17 Central Cad6.30.1 Cadaverous 0.0613 0.0588 0.0551 0.9646 0.9684 0.9603 0.0819 0.0749 0.0685 0.2759 18 Central Cad6.30.2 Cadaverous 0.0551 0.0536 0.0573 0.9603 0.9483 0.9626 0.0685 0.0694 0.0657 0.2965 19 Central Cad6.30.3 Cadaverous 0.0573 0.0560 0.0595 0.9626 0.4967 0.0053 0.0657 0.0664 0.0660 0.2568 20 Central Cad6.30.4 Cadaverous 0.0595 0.0820 0.0952 0.0053 0.0171 0.0226 0.0660 0.0866 0.0886 0.2428 21 Central Cad6.30.5 Cadaverous 0.0952 0.0753 0.0614 0.0226 0.0366 0.0445 0.0886 0.0655 0.0637 0.5398 22 Central Cr6.7.1 Crews 0.0614 0.0610 0.0566 0.0445 0.0357 0.0304 0.0637 0.0672 0.0722 0.2750 139
23 Central Cr6.7.2 Crews 0.0566 0.0637 0.0645 0.0304 0.0316 0.0390 0.0722 0.0770 0.0763 0.1541 24 Central Cr6.7.3 Crews 0.0645 0.0549 0.0502 0.0390 0.0506 0.0450 0.0763 0.0649 0.0552 0.2039 25 Central Cr6.7.4 Crews 0.0502 0.0483 0.0508 0.0450 0.0343 0.0346 0.0552 0.0667 0.0737 0.3622 26 Central Cr6.7.5 Crews 0.0508 0.0732 0.0981 0.0346 0.0388 0.0450 0.0737 0.0650 0.0711 0.2527 27 Central Cr7.13.1 Crews 0.0981 0.0939 0.0849 0.0450 0.0432 0.0353 0.0711 0.0827 0.0807 0.1670 28 Central Cr7.13.2 Crews 0.0849 0.0837 0.0774 0.0353 0.0253 0.0356 0.0807 0.0849 0.0850 0.1644 29 Central Cr7.13.3 Crews 0.0774 0.0876 0.0935 0.0356 0.0371 0.0264 0.0850 0.0713 0.0482 0.1646 30 Central Cr7.13.4 Crews 0.0935 0.0850 0.0838 0.0264 0.0299 0.0386 0.0482 0.0555 0.0813 0.2358 31 Central Cr7.13.5 Crews 0.0838 0.1069 0.1290 0.0386 0.0635 0.0828 0.0813 0.0845 0.0855 0.2324 32 Central Cr7.13.6 Crews 0.1290 0.0967 0.0605 0.0828 0.0547 0.0198 0.0855 0.0824 0.0621 0.2414 33 Central Cr7.13.8 Crews 0.0605 0.0577 0.0634 0.0198 0.0138 0.0208 0.0621 0.0544 0.0700 0.2308 34 Central Crys6.30.1 Crystal 0.0634 0.0663 0.0573 0.0208 0.0222 0.0188 0.0700 0.0841 0.0812 0.3290 35 Central Crys6.30.2 Crystal 0.0573 0.0562 0.0736 0.0188 0.4953 0.4960 0.0812 0.0548 0.0250 0.2390 36 Central Oe6.7.1 Eoff 0.0490 0.0623 0.0590 0.0311 0.0379 0.0386 0.0400 0.0608 0.0690 0.1963 37 Central Oe6.7.2 Eoff 0.0590 0.0605 0.0754 0.0386 0.0501 0.0683 0.0690 0.0626 0.0618 0.1661 38 Central Oe6.7.3 Eoff 0.0754 0.0616 0.0523 0.0683 0.0536 0.0398 0.0618 0.0680 0.0787 0.1506 39 Central Oe6.7.4 Eoff 0.0523 0.0803 0.1013 0.0398 0.0627 0.0736 0.0787 0.0704 0.0626 0.3119 40 Central Oe6.7.5 Eoff 0.1013 0.0708 0.0487 0.0736 0.0524 0.0424 0.0626 0.0602 0.0623 0.0567 41 Central Oe6.7.6 Eoff 0.0418 0.0555 0.0478 0.0425 0.0423 0.0237 0.0541 0.0705 0.0626 0.2972 42 Central Oe6.7.7 Eoff 0.0517 0.0516 0.0557 0.0330 0.0350 0.0467 0.0666 0.0657 0.0662 0.3904 43 Central Oe6.7.8 Eoff 0.0557 0.0662 0.0625 0.0467 0.0480 0.4781 0.0662 0.0633 0.0672 0.1813 44 Central Oe6.7.9 Eoff 0.0766 0.0485 0.0533 0.0489 0.9072 0.0242 0.0630 0.0713 0.0474 0.1495 45 Central Fall6.30.1 Falling Tree 0.0579 0.0602 0.0589 0.9960 0.9763 0.9497 0.0685 0.0687 0.0710 0.2150 46 Central Fall6.30.2 Falling Tree 0.0589 0.0657 0.0716 0.9497 0.9661 0.9773 0.0710 0.0544 0.0641 0.2954 47 Central Fall6.30.3 Falling Tree 0.0716 0.0512 0.0369 0.9773 0.4820 0.4856 0.0641 0.0689 0.0563 0.2626 48 Central Fall7.3.1 Falling Tree 0.0369 0.0502 0.0689 0.4856 0.5012 0.0426 0.0563 0.0579 0.0553 0.1523 49 Central G6.5.1 Gillespie 0.0596 0.0781 0.0766 0.0316 0.0536 0.0552 0.0549 0.0556 0.0665 0.3103 50 Central G6.5.2 Gillespie 0.0773 0.0605 0.0439 0.0544 0.0406 0.0279 0.0611 0.0623 0.0565 0.2161 140
51 Central G6.5.3 Gillespie 0.0439 0.0423 0.0606 0.0279 0.0161 0.4605 0.0565 0.0590 0.0689 0.2744 52 Central G6.5.4 Gillespie 0.0411 0.0800 0.0570 0.0023 0.9187 0.0135 0.0630 0.0748 0.0843 0.3757 53 Central GrOn6.29.1 Great Onyx 0.0812 0.0642 0.0667 0.9558 0.5039 0.0092 0.0733 0.0853 0.0833 0.3258 54 Central GrOn6.29.10 Great Onyx 0.0612 0.0728 0.0758 0.0113 0.5008 0.4974 0.0858 0.0886 0.0830 0.2671 55 Central GrOn6.29.11 Great Onyx 0.0758 0.0769 0.0770 0.4974 0.4976 0.9936 0.0830 0.0739 0.0790 0.2096 56 Central GrOn6.29.2 Great Onyx 0.0696 0.0555 0.0525 0.5019 0.4739 0.4740 0.0828 0.0802 0.0702 0.2866 57 Central GrOn6.29.3 Great Onyx 0.0525 0.0612 0.0677 0.4740 0.4977 0.4943 0.0702 0.0645 0.0760 0.3531 58 Central GrOn6.29.4 Great Onyx 0.0677 0.0669 0.0593 0.4943 0.4991 0.9910 0.0760 0.0793 0.0762 0.2765 59 Central GrOn6.29.5 Great Onyx 0.0593 0.0622 0.0650 0.9910 0.5052 0.0135 0.0762 0.0696 0.0647 0.1092 60 Central GrOn6.29.6 Great Onyx 0.0650 0.0486 0.0473 0.0135 0.4917 0.9845 0.0647 0.0601 0.0712 0.3248 61 Central GrOn6.29.7 Great Onyx 0.0473 0.0542 0.0506 0.9845 0.9737 0.9418 0.0712 0.0814 0.0817 0.2001 62 Central GrOn6.29.8 Great Onyx 0.0506 0.0568 0.0645 0.9418 0.4731 0.0150 0.0817 0.0814 0.0757 0.2504 63 Central GrOn6.29.9 Great Onyx 0.0645 0.0892 0.1094 0.0150 0.0166 0.0190 0.0757 0.0779 0.0800 0.2084 64 Central Hick7.3.3 Hickory 0.0387 0.0514 0.0363 0.0088 0.9739 0.9944 0.0641 0.0799 0.0773 0.2791 65 Central Hick7.3.4 Hickory 0.0438 0.0502 0.0658 0.9841 0.5085 0.0126 0.0786 0.0787 0.0842 0.2646 66 Central Hick7.3.5 Hickory 0.0640 0.0675 0.0544 0.0226 0.0026 0.0104 0.0801 0.0884 0.0855 0.2529 67 Central Hick7.3.6 Hickory 0.0610 0.0460 0.0488 0.0065 0.5023 0.5213 0.0870 0.0830 0.0743 0.3476 68 Central His6.29.1 Historical 0.0375 0.0601 0.0798 0.9941 0.0484 0.0586 0.0806 0.0681 0.0882 0.0289 69 Central His6.29.2 Historical 0.0700 0.0918 0.1026 0.0535 0.0613 0.0595 0.0781 0.0786 0.0726 0.1767 70 Central New6.29.1 New Discovery 0.0681 0.0626 0.0652 0.0134 0.0058 0.0058 0.0767 0.0750 0.0718 0.2868 71 Central New6.29.2 New Discovery 0.0660 0.0643 0.0811 0.0017 0.0098 0.0070 0.0762 0.0673 0.0819 0.2073 72 Central New6.29.3 New Discovery 0.0727 0.0755 0.0619 0.0084 0.0085 0.0117 0.0746 0.0821 0.0772 0.1524 73 Central New6.29.4 New Discovery 0.0619 0.0618 0.0688 0.0117 0.0172 0.0208 0.0772 0.0684 0.0708 0.1634 74 Central New6.29.5 New Discovery 0.0688 0.0563 0.0490 0.0208 0.0264 0.0311 0.0708 0.0557 0.0400 0.2399 75 Central Phil7.3.1 Phil 0.0509 0.0548 0.0522 0.4657 0.0265 0.0258 0.0594 0.0682 0.0768 0.1838 76 Central Phil7.3.2 Phil 0.0522 0.0629 0.0773 0.0258 0.0156 0.0120 0.0768 0.0736 0.0789 0.1349 77 Central Phil7.3.3 Phil 0.0773 0.0750 0.0794 0.0120 0.0164 0.0322 0.0789 0.0745 0.0796 0.1047 78 Central Phil7.3.4 Phil 0.0731 0.0856 0.0781 0.0170 0.0474 0.0536 0.0739 0.0852 0.0387 0.2196 141
79 Central Stan7.2.1 Stans Well 0.0614 0.0458 0.0407 0.0091 0.4997 0.9931 0.0546 0.0722 0.0696 0.4643 80 Central Stan7.3.1 Stans Well 0.0279 0.0535 0.0448 0.9872 0.9989 0.0058 0.0866 0.0526 0.0439 0.3074 81 Central Vici.29.5 Violet City 0.0592 0.0490 0.0467 0.0131 0.0078 0.4991 0.0847 0.0783 0.0752 0.1651 82 Central Vici.29.6 Violet City 0.0467 0.0690 0.0709 0.4991 0.9920 0.5040 0.0752 0.0831 0.0832 0.1597 83 Central Vici29.4 Violet City 0.0847 0.0572 0.0677 0.9885 0.0194 0.9984 0.0813 0.0851 0.0856 0.2009 84 Central Vici6.29.1 Violet City 0.0625 0.0682 0.0649 0.5089 0.5164 0.0294 0.0853 0.0856 0.0930 0.1784 85 Central Vici6.29.2 Violet City 0.0649 0.0635 0.0651 0.0294 0.0178 0.0211 0.0930 0.0911 0.0812 0.2621 86 Central Vici6.29.3 Violet City 0.0651 0.0742 0.0582 0.0211 0.0240 0.4966 0.0812 0.0830 0.0622 0.1307 87 Central Vici6.29.4 Violet City 0.0838 0.0325 0.0531 0.0168 0.9763 0.9911 0.0853 0.0390 0.0728 0.2056 88 Central Whi7.1.1 White 0.0428 0.0526 0.0516 0.9837 0.5053 0.4875 0.0559 0.0777 0.0718 0.1404 89 Central Whi7.1.2 White 0.0522 0.0511 0.0543 0.0194 0.9555 0.9787 0.0826 0.0609 0.0553 0.2634 90 Central YMCA7.4.1 YMCA 0.0527 0.0577 0.0676 0.9671 0.9773 0.4948 0.0581 0.0714 0.0862 0.1409 91 Central YMCA7.4.2 YMCA 0.0676 0.0562 0.0587 0.4948 0.0133 0.0200 0.0862 0.0698 0.0731 0.2396 92 Central YMCA7.4.3 YMCA 0.0587 0.0665 0.0600 0.0200 0.5067 0.4965 0.0731 0.0791 0.0685 0.2109 93 Central YMCA7.4.4 YMCA 0.0600 0.0540 0.0403 0.4965 0.4781 0.9652 0.0685 0.0669 0.0682 0.1619 94 Central YMCA7.4.5 YMCA 0.0403 0.0537 0.0498 0.9652 0.5063 0.0199 0.0682 0.0812 0.0693 0.4043 95 Central YMCA7.4.6 YMCA 0.0498 0.0396 0.0549 0.0199 0.4707 0.9653 0.0693 0.0563 0.0696 0.0961 96 Central YMCA7.4.7 YMCA 0.0549 0.0683 0.0777 0.9653 0.5092 0.0300 0.0696 0.0882 0.0978 0.2867 97 Central YMCA7.4.8 YMCA 0.0777 0.4387 0.7776 0.0300 0.3099 0.4536 0.0978 0.2911 0.5629 0.2420 98 Indiana E7.20.1 Lost River 0.0410 0.0406 0.0409 0.0195 0.0054 0.0118 0.0353 0.0283 0.0392 0.3338 99 Indiana E7.20.1c Lost River 0.0405 0.0414 0.0354 0.0049 0.0186 0.0120 0.0321 0.0464 0.0480 0.3269 100 Indiana E7.20.2 Lost River 0.0414 0.0354 0.0322 0.0186 0.0120 0.0217 0.0464 0.0480 0.0476 0.1835 101 Indiana E7.20.3 Lost River 0.0354 0.0322 0.0328 0.0120 0.0217 0.0146 0.0480 0.0476 0.0495 0.2013 102 Indiana E7.20.4 Lost River 0.0325 0.0348 0.0336 0.0182 0.0159 0.0205 0.0486 0.0496 0.0486 0.2070 103 Indiana E7.20.5 Lost River 0.0367 0.0305 0.0300 0.0172 0.0237 0.0278 0.0496 0.0477 0.0463 0.3375 104 Indiana E7.20.6 Lost River 0.0302 0.0428 0.0579 0.0258 0.5129 0.9960 0.0470 0.0605 0.0685 0.3444 105 Indiana Su7.20.1 Sullivan 0.0478 0.0442 0.0457 0.0151 0.0299 0.0388 0.0456 0.0460 0.0491 0.2921 106 Indiana Su7.20.2 Sullivan 0.0457 0.0509 0.0544 0.0388 0.0387 0.0453 0.0491 0.0535 0.0556 0.2213 142
107 Indiana Su7.20.3 Sullivan 0.0544 0.0517 0.0552 0.0453 0.0547 0.0651 0.0556 0.0548 0.0530 0.2236 108 Indiana Su7.20.4 Sullivan 0.0471 0.0633 0.0637 0.0584 0.0719 0.0401 0.0555 0.0506 0.0532 0.2459 109 Indiana Su7.20.5 Sullivan 0.0635 0.0596 0.0590 0.0560 0.0347 0.0411 0.0519 0.0353 0.0325 0.1919 110 Indiana Su7.20.6 Sullivan 0.0590 0.0624 0.0596 0.0411 0.0396 0.3973 0.0325 0.0683 0.0943 0.1999 111 Western BI7.11.1 Blue Moon 0.0667 0.1013 0.1195 0.0260 0.0377 0.0400 0.0831 0.0948 0.0965 0.1804 112 Western BI7.11.2 Blue Moon 0.1195 0.1132 0.1228 0.0400 0.0426 0.0402 0.0965 0.0798 0.0678 0.6128 113 Western BI7.11.3 Blue Moon 0.1228 0.1404 0.1627 0.0402 0.0469 0.0728 0.0678 0.0804 0.0986 0.5278 114 Western BI7.11.4 Blue Moon 0.1627 0.1275 0.0597 0.0728 0.0654 0.0275 0.0986 0.1008 0.0771 0.5666 115 Western BI7.11.5 Blue Moon 0.0681 0.0669 0.0665 0.0531 0.0233 0.0287 0.0903 0.0835 0.0828 0.5873 116 Western BI7.11.5 Blue Moon 0.0883 0.0311 0.0364 0.0504 0.0046 0.0111 0.1031 0.0510 0.0428 0.2059 117 Western Bu17.14.4 Bull Creek 0.0862 0.0865 0.0813 0.0368 0.0487 0.0476 0.0954 0.0840 0.0857 0.1430 118 Western Bu17.14.5 Bull Creek 0.0839 0.0751 0.0716 0.0482 0.0362 0.0410 0.0849 0.0813 0.0822 0.1707 119 Western Bu17.14.6 Bull Creek 0.0716 0.0771 0.0761 0.0410 0.0523 0.0368 0.0822 0.0807 0.0758 0.2856 120 Western Bu17.14.7 Bull Creek 0.0761 0.0724 0.0646 0.0368 0.0261 0.0321 0.0758 0.0778 0.0638 0.1985 121 Western Bu17.14.8 Bull Creek 0.0646 0.0739 0.0931 0.0321 0.0438 0.0590 0.0638 0.0640 0.0807 0.2530 122 Western Bu17.14.9 Bull Creek 0.0931 0.0772 0.0803 0.0590 0.0554 0.0439 0.0807 0.0770 0.0847 0.1372 123 Western Bu7.14.1 Bull Creek 0.0983 0.0752 0.0715 0.0436 0.0416 0.0416 0.0840 0.0582 0.0592 0.1474 124 Western Bu7.14.12 Bull Creek 0.0591 0.1014 0.0952 0.0424 0.0454 0.0418 0.0706 0.0988 0.0691 0.1864 125 Western Bu7.14.2 Bull Creek 0.0553 0.0877 0.0901 0.0414 0.0418 0.0471 0.0473 0.0710 0.0874 0.1831 126 Western Bu7.14.3 Bull Creek 0.0889 0.0574 0.0209 0.0445 0.5192 0.4965 0.0792 0.0665 0.0454 0.1303 127 Western IG7.10.1 Iron Gate 0.1026 0.1058 0.1017 0.0595 0.0546 0.0592 0.0726 0.0889 0.0976 0.2656 128 Western IG7.10.2 Iron Gate 0.1017 0.0890 0.0842 0.0592 0.0570 0.0423 0.0976 0.0900 0.0867 0.2312 129 Western Ig7.10.3 Iron Gate 0.0850 0.0835 0.0711 0.0498 0.0348 0.0505 0.0864 0.0870 0.0762 0.2164 130 Western IG7.10.3 Iron Gate 0.0835 0.0711 0.0783 0.0348 0.0505 0.0402 0.0870 0.0762 0.0895 0.2179 131 Western J7.10.1 Jail 0.0747 0.0791 0.0761 0.0454 0.0433 0.0472 0.0829 0.0849 0.0803 0.2343 132 Western J7.10.2 Jail 0.0800 0.0722 0.0923 0.0464 0.0480 0.0646 0.0802 0.0804 0.0753 0.1887 133 Western J7.10.3 Jail 0.0822 0.0807 0.0794 0.0563 0.0631 0.0565 0.0779 0.0749 0.0826 0.2492 134 Western J7.10.4 Jail 0.0794 0.0807 0.0866 0.0565 0.0494 0.0617 0.0826 0.0973 0.1007 0.2292 143
135 Western J7.10.5 Jail 0.0717 0.1015 0.1058 0.0474 0.0760 0.0657 0.1039 0.0976 0.1158 0.2024 136 Western J7.10.6 Jail 0.1036 0.0937 0.0719 0.0709 0.0632 0.0528 0.1067 0.1007 0.0771 0.1681 137 Western J7.10.7 Jail 0.0719 0.0673 0.0870 0.0528 0.0356 0.0407 0.0771 0.0740 0.0919 0.2315 138 Western J7.10.8 Jail 0.0724 0.1017 0.0693 0.0262 0.0552 0.0411 0.0795 0.1044 0.0938 0.1318 139 Western J7.10.9 Jail 0.0855 0.0650 0.0667 0.0482 0.0490 0.0568 0.0991 0.0855 0.0819 0.1752 140 Western Jan7.9.1 January-Stansbury 0.0667 0.0674 0.0612 0.0568 0.0449 0.0340 0.0819 0.0758 0.0709 0.2129 141 Western Jan7.9.2 January-Stansbury 0.0612 0.0737 0.0781 0.0340 0.0378 0.0477 0.0709 0.0772 0.0828 0.1522 142 Western Jan7.9.3 January-Stansbury 0.0781 0.0720 0.0724 0.0477 0.0486 0.0363 0.0828 0.0786 0.0684 0.1609 143 Western Jan7.9.4 January-Stansbury 0.0724 0.0761 0.0874 0.0363 0.0269 0.0310 0.0684 0.0828 0.0906 0.2712 144 Western Jan7.9.5 January-Stansbury 0.0874 0.0813 0.0794 0.0310 0.0396 0.0415 0.0906 0.0752 0.0693 0.2174 145 Western Jan7.9.6 January-Stansbury 0.0794 0.0767 0.0770 0.0415 0.0518 0.0360 0.0693 0.0700 0.0470 0.3079 146 Western Jan7.9.7 January-Stansbury 0.0770 0.0693 0.0458 0.0360 0.5040 0.9925 0.0470 0.0430 0.0560 0.2071 147 Western S7.11.1 Survivalist 0.0819 0.0750 0.0686 0.0505 0.0378 0.0426 0.0620 0.0616 0.0622 0.1940 148 Western S7.11.2 Survivalist 0.0686 0.0622 0.0614 0.0426 0.0346 0.0091 0.0622 0.0456 0.0546 0.1991 149 Western Th7.11.1 Third 0.0596 0.0637 0.0678 0.3973 0.3939 0.0117 0.0943 0.0971 0.0949 0.3339 150 Western Th7.11.2 Third 0.0678 0.0662 0.0675 0.0117 0.0207 0.0436 0.0949 0.0848 0.0786 0.3829 151 Western Th7.11.3 Third 0.0675 0.0754 0.0784 0.0436 0.0558 0.0515 0.0786 0.0876 0.0843 0.2297 152 Western Th7.11.4 Third 0.0784 0.0672 0.0592 0.0515 0.0273 0.0131 0.0843 0.0772 0.0847 0.1635
144
Table S2. Average range-wide PST values produced from a sensitivity analysis using a range of additive genetic variance and heritabilities.
Orange PC1 g=1 g=.1 g=.01 h=1 0.182 0.035 0.004 h=0.75 0.212 0.045 0.005 h=.5 0.257 0.063 0.008 h=0.25 0.338 0.104 0.016 h=.1 0.442 0.182 0.035 Spot PC1 g=1 g=.1 g=.01 h=1 0.252 0.080 0.024 h=0.75 0.283 0.093 0.028 h=.5 0.329 0.116 0.034 h=0.25 0.408 0.165 0.049 h=.1 0.502 0.252 0.080 Ratio g=1 g=.1 g=.01 h=1 0.139 0.021 0.002 h=0.75 0.166 0.028 0.003 h=.5 0.209 0.040 0.005 h=0.25 0.288 0.071 0.009 h=.1 0.391 0.139 0.021
145
APPENDIX A
DEVELOPMENT AND CHARACTERIZATION OF 19 MICROSATELLITE LOCI IN
THE CAVE SALAMANDER, EURYCEA LUCIFUGA (PLETHODONTIDAE)
Abstract
Subterranean environments are home to a multitude of species of conservation concern, since most cave-restricted organisms are endemic to a small range. Because of their conservation status and the physical difficulty of studying cave-restricted species, research on the evolutionary history and population dynamics of cave species is challenging. We have developed and characterized novel microsatellite markers using transcriptomic methods for the cave salamander, Eurycea lucifuga. Since this species is commonly found in caves across a wide expanse of the United States, these markers will facilitate studies of the cave habitat in a system that is relevant and tractable.
146
Methods and Results
The cave salamander, Eurycea lucifuga, is cave-dwelling Plethodontid species inhabiting a large range which extends across the Central and Eastern United States
(Petranka 1998). In contrast with many vertebrate species that inhabit caves, which are often endemics of conservation concern, E. lucifuga is commonly found across its range in far greater numbers. The species therefore is well suited to studies of the population dynamics and evolutionary history of a unique and conservationally important habitat.
The development of molecular resources, e.g. microsatellite markers, in E. lucifuga will allow more detailed study of ecological and evolutionary processes such as gene flow within and among cave systems, evolutionary dynamics of morphological change, and phylogenetic exploration of the relationship between E. lucifuga and closely related species.
Tail clips were taken from two individuals collected from different localities in
Southwestern Virginia in Spring 2014 (Smokehole Cave 37.31402N 80.50894W, and
Blankenship Blowhole 37.3135N 80.84987W; VADGIF permit #042984), and immediately preserved in RNAlater (Life Technologies). Whole sample RNA was extracted using a Qiagen RNEasy Mini Kit (Qiagen, Valencia, CA) following manufacturer suggested protocols and sent to the Genomic Core Facility at the University of Virginia’s Department of Biology for the construction of individual paired-end, non- normalized cDNA libraries using a NEBNextmRNA library Prep Master Mix Set (New
England Biolab Inc., Ipswich, MA). Individual libraries were given barcodes and pooled, and sent to Genewiz (GENEWIZ Inc., South Plainfield, NJ) and sequenced on an
Illumina HiSeq 2500 using a PE 2x100bp format. Sequences for each sample were 147 assembled using the program Trinity v.11-10-2013 (Grabherr et al. 2011), and a summary of the sequencing results for each library can be found in Table 1. A custom
BLAST search was used to locate regions of the mitochondrial genes cytb and nd2 sequenced previously from E. lucifuga in order to confirm the identity of the samples.
Putative microsatellite loci were identified in each group of transcripts using the program msatcommander v. 1.0.8 (Faircloth 2008), specifying that the program search for repeats of trinucleotides or greater with at least five repeats. We used a BLAST searches to confirm the presence of primer sequences in both samples, and discarded primers that were not present in both group of transcripts.
Following the method of Schuelke (2000), we screened each primer set for amplification and polymorphism using m13 fluorescently labeled tags. Each test was performed using eight samples collected from across the range, and confirmatory sequencing was done using fragment analysis on an ABI 3130 Sequencer (Life
Technologies). Primer pairs that amplified in all samples and were polymorphic were ordered with fluorescent tags incorporated into each forward primer, and multiplexed.
Thermocycler conditions used throughout the screening process are: 94° for 15:00, 40 cycles of 94° for 0:30, 60° for 1:30, and 72° for 1:30, and finally 72° for 10:00. After identifying putative microsatellite loci and using BLAST to validate them, we were left with 220 primer pairs to screen. Twenty-four of these amplified in all individuals and were polymorphic, (>2 alleles) and these were multiplexed into six multiplexes, each containing four loci (Table 2).
The variability of these multiplexes was tested using 261 samples of Eurycea lucifuga collected from 49 populations across its range (Table 3). Fragment analysis for this test 148 was performed at the DNA Analysis Facility on Science Hill at Yale University (New
Haven, CT). Allele number at each locus (k) ranged from 2-7 (Table 2). Observed and expected heterozygosity (Ho and Hs) were estimated using GenoDive v.2.0b27
(Meirmans & Van Tienderen 2004), and are shown in Table 2. When analyzed in
GenoDive v. 2.0b27, no loci significantly deviated from HWE within populations (Table
4). All analyses were run using 999 permutations. These markers will enable further research on population and evolutionary dynamics in Eurycea lucifuga and potentially close relatives.
References Faircloth BC (2008) MSATCOMMANDER: detection of microsatellite repeat arrays and automated, locus-specific primer design. Molecular ecology resources 8, 92-94. Grabherr MG, Haas BJ, Yassour M, et al. (2011) Full-length transcriptome assembly from RNA-Seq data without a reference genome. Nat Biotech 29, 644-652. Meirmans PG, Van Tienderen PH (2004) GENOTYPE and GENODIVE: two programs for the analysis of genetic diversity of asexual organisms. Molecular Ecology Notes 4, 792-794. Petranka J (1998) Salamanders of the United States and Canada Smithsonian Institution Press, Washington. Schuelke M (2000) An economic method for the fluorescent labeling of PCR fragments. Nature biotechnology 18, 233-234.
149
Tables
Table 1. Summary of the sequencing results and assembly of the two paired-end libraries. Library Number of Q>30 Total Median contig Total reads transcripts length assembled bases BLB 128 mill 89.96% 112,449 355bp 73,398,489 Smoke 300 mill 90.26% 92,594 362bp 59,509,037
Table 2. Microsatellite markers developed for Eurycea lucifuga, reported with primer sequence information as well as multiplex configuration and motif. We characterized the number of alleles at each locus (k), as well as observed (Ho) and expected (Hs) heterozygosity. Marker name Forward primer Reverse Primer Plex Motif k Ho Hs E_luc_238 ATGGCTGCGCTTTCTTGTAC CTCTGTACAGGAGACGGGTG 1 AAG 3 0.009 0.009 E_luc_915 TGCCGAAAGTTGCAGTGAAG CGCATCGTCATCTGCAGAAG 1 ATC 4 0.04 0.048 E_luc_1405 ACTGAGCAAACTTCGCATGG TGTCCAGATGCCTCTACAGC 1 ACCT 5 0.092 0.099 E_luc_1259 ACAGCTTGCTTACTTGGTGC AAGGGAACAAGGCTCAGAGG 2 AAG 4 0.473 0.519 E_luc_1375 ACAAGCTCCATTTGCACGAG GTGGTAGCCCTGGTTCTAGG 2 AACC 5 0.13 0.175 E_luc_1284 GGTCTTTGTCAGCAGTGCAG CCGAGGGCCTAAGTCTAACC 2 AAGC 5 0.194 0.273 E_luc_423 GGATGAAGAAGGGTACTGCG GCTGACTCTTGCAGACTGTG 3 ACC 3 0.198 0.204 E_luc_1055 TGTGGTTGTATGCTTATCAGGC TTCTGTGTGCTCAAGGAGATG 3 AAT 3 0.094 0.205 E_luc_433 TGGAAAGGAAGCCAAAGTCAC GTGCCAAATCCCTCTGCATC 4 AAT 4 0.107 0.267 E_luc_971 CAGCCACAATCCAAGAACCC AAGCCGGAATAGTAGAGCCG 4 ATC 7 0.431 0.543 E_luc_240 TGCTATGACCTCTGGCATCC AAGTTCTCCAGAGGCCTTGG 4 AAAT 2 0.174 0.25 E_luc_2440 GCAGCAGAAACAAGGACTGG CCAGTCTGACAGTGCGGG 4 AGGG 7 0.211 0.286 E_luc_371 GTATGTGTGCACTGCGAGAG TCAGTGGCTTGGATCTGGTG 5 ACC 5 0.033 0.04 E_luc_2319 ATCAACGTTCTGAATGCGCC TGCACTGAACTAGGAGGGAC 5 AAT 5 0.361 0.381 E_luc_961 TGTTGCAAAGTTCTGGTCGG CGTGCTTTACTTCCTTGGCC 5 ACC 4 0.206 0.215 E_luc_808 CCCAGAACATGCACAACCAG TAGCGGCTGGAAGAAGGATC 6 AGC 5 0.655 0.499 E_luc_2336 TTTCATGGCTGCTTGTACCC ACATACTACAACTCGAGGTGC 6 AAAT 3 0.04 0.046 E_luc_566 AGGGTTTAACTGCTGAAGGG GCAAATCTCAGCCGTGTCTC 6 AAAT 4 0.04 0.198 E_luc_2121 CCCTCCCTGTGCTTACTCTG ACGATCTGACCTGATGACCG 6 AATT 2 0.014 0.013
150
Table 3. Location and collection information of each cave locality from which samples were taken. N indicates the number of sampled individuals at each locality. General regions only are provided, due to cave conservation and safety concerns, but geographic coordinates are available upon request. Cave Name State N Cave Name State N Bankley/Fairground IN 9 YMCA KY 8 Donnehue IN 5 Bull Creek MO 9 Roberts IN 4 Crighton Spring MO 5 Robinson Ladder IN 4 Blue Moon OK 2 Sullivan IN 6 Iron Gate OK 3 The Lost River IN 4 Jail OK 9 Adwell KY 1 January-Stansbury OK 7 Big Hollow Cave KY 7 Survivalist OK 2 C2 KY 7 Third OK 4 Cadaverous Cave KY 5 Crews TN 4 Crystal Cave KY 2 Eoff TN 6 Falling Tree Cave KY 6 Gillespie TN 4 Great Onyx Cave KY 11 Lost Puddle TN 1 Hickory Cabin Cave KY 6 Mull Prowell TN 6 Left Eye Cave KY 1 Pompie TN 2 Little KY 2 Blankenship Blowhole VA 6 Mammoth Cave KY 2 Byrd's Water VA 10 Natural Bridge Cave KY 1 Smokehole VA 6 New Discovery Entrance KY 11 Tawney's VA 3 Phil Cave KY 4 Borehole WV 4 Stan KY 2 Buckeye Creek WV 1 Sturgeon Cave KY 1 Higganbotham WV 4 Unknown Cave (Austin Ent.) KY 2 Mann WV 3 Violet City Entrance KY 6 Spring WV 1 White Cave KY 2
151
Table 4. Hardy-Weinberg estimations for each locus within each population. Significant departures from Hardy-Weinberg are indicated in bold. Population 238 915 1405 1259 1375 1284 423 1055 433 971 240 2440 371 2319 961 808 2336 566 2121 Multi-locus
Adwell 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 --- Austen --- 0 0 0 0 0 --- 1 0.5 0 0 0 --- 0 0 0 ------0.25 Buckeye_Creek 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 --- Bankley ------0.091 0.216 --- -0.231 -0.212 0 --- -0.116 0.158 -0.067 --- 0 -0.164 0.048 --- 0.536 --- 0.034 Big ------0 0.321 0 -0.125 0 1 0.647 0.419 0.368 -0.358 --- 0.213 -0.125 -0.2 --- 1 --- 0.214 Blue_Moon ------0 --- 0 ------1 --- 0 ------1 ------0.5 Black 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 --- Blankenship ------0.062 ------0.5 0.524 ------0.538 0.333 ------0.141 Borehole ------0 ------0.5 ------0.5 --- 1 0 0.25 Bull_Creek ------0.333 -0.44 ------0.59 -0.514 ------0 -0.067 --- -0.778 --- 0.636 --- -0.173 Byrd --- 0.64 1 0.237 0.64 1 --- 1 0.673 0.308 0 0.64 0.654 0.64 0.5 -0.191 0.386 1 0 0.463 C2 ------0 -0.429 1 -0.111 0 -0.091 0.5 0 0.727 0.41 --- 0.368 0.122 0.294 ------0.252 Cadaverous ------0.5 1 0.636 -0.143 1 0.5 -0.333 -0.6 1 ------0.2 -0.067 ------0.197 Crews ------0 -1 0 ------1 ------0 0 --- 0 ------0.286 Crystal ------1 --- 0 0 --- 0 --- 0 ------1 0 ------0.125 Crighton ------0 -0.143 1 ------1 0.273 --- 0.6 ------1 0 ------0.226 Culver 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 --- Donnehue ------0.333 --- 0.6 -0.143 0 1 0.75 0.6 ------0.04 --- 1 --- 0.432 Lost ------0.282 0 0.487 -0.154 ------0.25 -0.25 ------0.111 0.211 --- 0.643 --- 0.046 Fall ------0 0.318 0 -0.053 --- 0.615 0 0.211 0.615 0.211 --- 0.024 -0.111 0.118 ------0.204 Gillespie --- 0 -0.091 --- -0.091 --- 1 ------0.182 --- 0.571 0 0 0 0.143 ------0.2 Great_Onyx ------0 -0.033 0 0.304 0.268 0.574 0.223 0.459 0.259 0.231 --- 0.32 -0.026 -0.19 0 1 --- 0.237 Higganbotham ------1 0 ------0.6 --- 0 ------0 ------0 Higganbotham2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 --- Hickory ------0 -0.163 --- -0.143 1 -0.111 --- 1 -0.25 -0.053 --- 0.259 0 0.25 ------0.128 Historical ------0 0 0 0 0 --- 0 0.5 1 -1 --- -0.333 0.5 -0.333 ------0.154 152
Iron_Gate ------0 ------0 --- 1 ------0 -1 --- 1 --- 0.333 Jail --- -0.067 --- -0.032 -0.2 ------1 -0.366 --- 0 --- 0 1 -0.8 --- 0.644 --- 0.028 January --- -0.091 --- 0 1 ------0.5 -0.714 ------0.2 --- -1 --- 1 --- 0.016 Left 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 --- Little ------0 --- 0 0 --- 0.5 1 1 0 --- 0.5 --- 0 ------0.545 Puddle 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 --- Mann ------1 ------0 ------0.2 ------1 --- 0.6 --- 0.263 Natural 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 --- New ------0 0.151 --- 0.64 0 0.31 0.5 0.474 0.259 -0.163 --- 0.124 0.268 -0.149 ------0.209 Eoff --- 0 --- 0.706 -0.053 -0.053 -0.111 -0.429 --- 0.091 --- 0 0 0.362 0.062 -0.395 --- 0 --- 0.043 Pagoda --- 0 --- -0.333 --- 0 --- 1 1 --- 0 ------0.333 -0.143 0 ------0.216 Phil ------0 0.294 --- 0 -0.2 1 1 0 0 -0.091 --- -0.412 0 0 ------0.165 Pompie ------0 1 --- 0 0 ------0.5 --- 0 0 -1 --- 0 --- 1 --- 0.308 Prowell -0.154 --- -0.111 -0.667 -0.25 0.615 -0.429 0.394 0.286 0.286 --- -0.081 --- -0.19 --- -0.25 --- 1 0 0.024 Roberts ------0.571 --- 1 -0.2 1 --- 0.667 0 ------0.5 --- 0 --- 0.362 Robinson ------0.059 --- 0.294 -0.2 --- 1 0.625 0 ------0.2 -0.412 ------0.174 Survivalist ------1 0 ------1 --- 1 --- 0 --- -1 ------0.167 Silver ------0.391 0.615 1 -0.25 0.333 0.5 0 0.615 -0.053 --- 0.302 -0.154 0 ------0.21 Smoke ------0.706 ------0 ------0.211 ------0.25 ------0.18 Spring 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 --- Stan ------0 -1 --- 1 --- 0 ------0 0 --- -0.333 --- 0 ------0.053 Sturgeon 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 --- Sullivan ------0.081 --- -0.429 -0.111 --- 1 0.211 0.062 ------0.111 0 --- 0.706 --- 0.149 Tawney ------1 ------0 ------0.2 ------0.111 0 ------0.314 Third ------0 ------0.2 ------0.2 --- -1 --- 1 --- -0.071 Violet ------0.524 0 -0.111 --- 0.333 0.5 --- -0.25 0.062 0 -0.081 -0.111 ------0.062 --- 0.121 White ------0 --- 0 ------0 -1 0.5 --- 1 0 ------0.222 YMCA ------0.111 --- -0.077 -0.167 -0.077 0.65 0.192 0.741 -0.273 --- 0.079 -0.077 -0.105 ------0.089 153
Overall 0.011 0.165 0.077 0.087 0.257 0.29 0.045 0.541 0.597 0.197 0.303 0.261 0.164 0.051 0.042 -0.31 0.094 0.8 0.114 0.18